by david927 on 5/25/25, 7:36 PM with 1087 comments
by creakingstairs on 5/26/25, 12:37 AM
At the moment, the flow goes like this:
1. During the week, write posts for things that have happened.
2. Posts can be assigned to groups. (family, friends etc).
3. At the end of the week (or month), the app automatically creates a newsletter for each group by pulling posts assigned to each group. Add some final touches yourself and send it off!
4. Every newsletter will come with a link to download all images.
I'm trying to design it to be as old people friendly as possible which meant making the experience as simple as it can get. This made me settle on email newsletters. Emails are ubiquitous, have been around and will be around for a while. It's easy to sign up and things are just pushed to you instead of having to go to another app.
Another thing I want is multilingual support as my family is Korean my in-laws are not.
I'm hoping to get an MVP working this week and get some testing done with my own parents and in-laws.
by acidburnNSA on 5/25/25, 8:50 PM
- Nuclear Reactor Starter Kit --- an open source set of procedures, processes, templates, and maybe even some IT advice that should help newcomers start companies with nuclear quality assurance programs easily and quickly while also making a new format in which nuclear companies can share lessons learned in efficiency.
- Reactor Database --- similar to the iaeas PRIS but focused on reactor development rather than power reactors. Will include nuclear startup company tracking with details gleaned from statements and maybe extrapolated where necessary from simple simulations. Will include things like fuel cost and licensing progress. This way people can more easily separate vaporware from real nuclear, and keep track of promises vs delivery.
by z3ugma on 5/26/25, 1:54 AM
by rorylaitila on 5/25/25, 8:56 PM
I have tens of thousands of ads in the collection and it would take me many lifetimes to complete, but I've been using AI to extract and catalog the meta data. I can get through about 100 ads/day this way.
One of my favorite ads, a computer from 1968 that "answers riddles": https://adretro.com/ads/1968-digi-comp-digi-comp-1-table-top...
by aantix on 5/25/25, 9:36 PM
It allows us to control the algorithm. It’s all LLM translating to YouTube search queries under the hood.
Visually it looks the same. The suggested videos come from predefined buckets on topics they love.
E.g. 33% fun math, 33% DIY engineering, 33% creative activities.
Video recommendations that have a banned word in the title/desc don't get displayed e.g. MrBeast, anything with Minecraft in it, never gets surfaced.
For anyone interested in using it, send me an email.
I'll put you on my list. And you can contribute ideas to our community Google Doc.
jim.jones1@gmail.com
by BSTRhino on 5/25/25, 10:26 PM
A game engine that lets you code multiplayer games without coding the multiplayer! My idea was to put multiplayer into the fabric of the programming language itself. This allows the engine to automatically turn your game into a multiplayer game, without you needing to learn anything about networking or synchronization. I imagine there are lots of people who have the talent and creativity to create a multiplayer game but don't have the interest or patience in learning how to code multiplayer, and so that's who this is for!
I've been working on this for 3 years and there were lots of tricky parts rolling back and deterministically executing a whole programming language, but it's working now! My next phase is to increase the breadth of features so better games can be made with it!
by funvill on 5/26/25, 4:36 AM
A random button in a park with a countdown timer.
Instructions:
- Press the button to reset the 24-hour countdown timer.
- If the timer ever reaches zero, the project ends and the project will be removed.
- To keep the project alive — press the button
Its been running for 56 days with 820 button presses.
Write up: https://blog.abluestar.com/projects/2025-counterproductive/
Stats page https://blog.abluestar.com/other/counterproductive.html
by ayaros on 5/25/25, 10:06 PM
by jesse__ on 5/26/25, 3:43 PM
https://github.com/scallyw4g/bonsai
I also wrote a metaprogramming language which generates a lot of the editor UI for the engine. It's a bespoke C parser that supports a small subset of C++, which is exposed to the user through a 'scripting-like' language you embed directly in your source files. I wrote it as a replacement for C++ templates and in my completely unbiased opinion it is WAY better.
by catchmeifyoucan on 5/25/25, 10:20 PM
I'm designing everything from the chassis to the software OS. E-ink has its own design constraints. I'm building 5 apps for it: a browser, reader, mail, writer and code editor. It's still a ways to go. Here's a picture of what I have so far:
by michelangelodev on 5/25/25, 10:43 PM
I was a YC founder in 2006 and still do software engineering and data science full-time, but on the side I also do Christian apologetics, helping fellow engineers/scientists/mathematicians seek answers to life's deepest questions.
Some cool articles for the HN crowd:
- My interview of Evan O'Dorney, a three-time Putnam Fellow and two-time IMO gold medalist, who converted to Catholic Christianity: https://www.saintbeluga.org/veritas-part-i-conversion-of-a-p...
- In-depth scientific overview of Eucharistic miracles: https://www.saintbeluga.org/eucharistic-miracles-god-under-t...
- Conversion testimony by the Chief Scientist at NASA JPL: https://www.saintbeluga.org/veritas-part-iii-bellows-of-aqui...
by pasxizeis on 5/26/25, 9:33 AM
So I thought: what if, when you opened the PR, you had a tool that actually executed your migrations, saw what locks they acquired (in runtime) and then reported it as a comment to your PR? This would provide you with yet another data point you could use to reason about whether your migration is safe to deploy or not. For example:
ALTER TABLE users ADD COLUMN email TEXT:
acquired AccessExclusiveLock on relation `users`
It's in early-stage and very much a PoC yet, but I think there's merit in the idea.
https://github.com/agis/pglockanalyze
P.S. Also, an excuse to learn Rust - so any feedback from Rust ppl more than welcome
by carpo on 5/25/25, 11:16 PM
In the app you pick a folder with videos in it and it stores the path, metadata, extracts frames as images, uses a local whisper model to transcribe the audio into subtitles, then sends a selection of the snapshots and the subtitles to an LLM to be summarised. The LLM sends back an XML document with a bunch of details about the video, including a title, detailed summary and information on objects, text, people, animals, locations, distinct moments etc. Some of these are also timestamped and most have relationships (i.e this object belongs to this location, this text was on this object etc). I store all that in a local SQLLite database and then do another LLM call with this summary asking for categories and tags, then store them in the DB against each video. The App UI is essentially tags you can click to narrow down returned videos.
I plan on adding a natural language search (Maybe RAG -- need to look into the latest best way), have half added Projects so I can group videos after finding the ones I want, and have a bunch of other ideas for this too. I've been programming this with some early help from Aider and Claude Sonnet. It's getting a bit complex now, so I do the majority of code changes, though the AI has done a fair bit. It's been heaps of fun, and I'm using it now in "production" (haha - on my PC)
by jamesfly on 5/26/25, 6:07 AM
It’s the biggest and dirtiest dataset I’ve ever worked with, so it’s been interesting to figure out practical solutions to run things fast and generalize cleaning tasks. Of course it’ll be impossible to get every case (I can only match about half of the state licenses to national records at the moment), so I’ll have to figure out a user-edit/consensus system for the rest.
by ml- on 5/25/25, 9:58 PM
Decided to do an extended sabbatical after being part of one of the many tech layoffs the last years, and I'm thus working on things I like, instead of things that pay..
Collecting and cataloging craft beer venues from around the world, at https://wheretodrink.beer Still a WIP, and it's not trying to be the most extensive list, but I want it to be a substantial list. Once I reach a certain maturity in the data I'll probably look to spawn minor projects off from the data set.. have a couple ideas already that I'll just keep to my self for now :D
I also had a set of left over domains relating to beer that I'm offering up for use with BlueSky handles, and beer related link pages at https://drnk.beer - a bit on the back burner.
by mingodad on 5/25/25, 9:36 PM
The idea is to improve the tooling to work with grammars, for example generating railroad diagrams, source, stats, state machines, traces, ...
On both of then select one grammar from "Examples" then click "Parse" to see a parse tree or ast for the content in "Input source", then edit the grammar/input to test new ideas.
There is also https://mingodad.github.io/plgh/json2ebnf.html to generate EBNF for railroad diagram generation form tree-sitter grammars.
Any feedback, contribution is welcome !
by greentec on 5/25/25, 10:55 PM
I analyzed 7 years of Armorgames.com data (999 games) to understand web gaming market trends.
Key findings that might interest fellow developers:
User standards are rising: Average ratings dropped from 7.02 (2018) to 6.45 (2025), but the percentage of high-quality games (8.5+ rating) actually increased from 12.3% to 14.7%. This suggests quality polarization rather than overall decline.
Genre trends: Rising: Idle games, Strategy, RPGs (deeper gameplay mechanics) Declining: Traditional arcade/action games Stable: Puzzle and Adventure (web gaming staples)
Innovation wins: The highest-rated "hidden gems" all had one thing in common - innovative mechanics rather than genre variations. Games like "Detective Bass: Fish Out of Water" (9.3 rating) and "SYNTAXIA" (9.1 rating) show originality still pays off.
Market maturation: The correlation between rating and popularity is surprisingly weak (0.126), suggesting quality ≠ virality. However, play count strongly correlates with favorites (0.712).
by benhoyt on 5/25/25, 8:47 PM
by heliographe on 5/25/25, 10:55 PM
I had fun with the interface - it's themeable, and inspired by classic cameras: lets you quickly switch between full auto/half auto/full manual modes with dedicated dials.
Going to add more features in the coming months, but the #1 focus is keeping it super simple and blazing fast.
Given that virtually all processing pipelines these days stack multiple shots to create a photo, as far as I'm aware this is the only way of getting a "traditional" single-exposure photo on iPhone, where the shutter speed is actually meaningful.
There are other camera apps that support Bayer RAW capture, but those support a bunch of other formats, and you probably don't want Bayer RAW for most of your shots anyways, so for my own workflow it's better to have a dedicated app that I can launch really quickly rather than tap around in menus.
by nsarrazin on 5/26/25, 2:05 PM
The whole thing works pretty well, it's built using Godot & a custom Rust extension with the 6502 emulator & assembler. It can even be exported to the web so you can write your code using a CodeMirror editor and send it to the game in the browser.
Now, as expected, the hard part is making it fun. Just giving players an empty code editor and a manual for the sensors/actuators doesn't seem great, and there's also no goal for now. I'd like to take inspiration from Shenzhen I/O to build a sense of progression by tackling more and more difficult control challenges maybe?
It also supports running multiple computers at once so I think there's some fun potential for a "multiplayer" mode where you try to achieve some objective (scanning X, mining Y) against other players by sending spacecrafts with your code on them and all the crafts would run on one server. Anyway I'm curious if people would be interested in such a game!
by fizzbuzz07 on 5/26/25, 9:34 AM
by T0Bi on 5/25/25, 11:07 PM
Everything from farm related stuff (water, food, shelter, etc.) to self-sufficiency (solar, etc.) to real time monitoring (which cameras, affecting power supply).
Who knows if it'll ever happen, but just planning everything in detail is a lot of fun. Especially with weird regulatory constraints where I'm living, there's a lot to watch out for.
Example: Solar panels at >3m height need building permits. Snow in winter means panels should be set up at a specific angle. So my initial plan of putting the panels on my 2.5m high carport doesn't work. Either lower carport, lower angle, different place or getting a building permit.
by colinnordin on 5/26/25, 11:27 AM
So I started building something simple: a tool that lets me turn highlights into flashcards with as little friction as possible.
Just select text on your iPhone, share it with the app, and it creates a flashcard using AI — a Q&A pair and a short summary. You can browse cards in the app, or show them on your Home Screen, Lock Screen, or watchface of your Apple Watch.
This is my first iOS app, and building it has been a great learning experience. I’m using Supabase for the backend which have been mostly great.
Check it out: https://komihag.com
by bengold14 on 5/25/25, 10:39 PM
I've seen some pretty fun novel use cases, such as (multiple!) people using it to pick out glasses, wedding invites & so on.
I recently completed a leaderboard function that cross compares photos from different tests using Claude, which was really impressive and scared me for my day job..
by ccvannorman on 5/25/25, 11:31 PM
A 3D game to help students in grades 5-8 learn Arithmentic, Fractions, Geometry, and Algebra.
50% or more of middle school students experience math anxiety, and it's no wonder that so many people grow up believing, "I'm not a math person." Math can be incredibly fun and beautiful if approached and experienced the right way. Mathbreakers is a vibrant, interactive world where all game mechanics are built on intrinsic mathematical properties, so simply by playing the game, a foundation of understanding of those concepts is built.
We're doing early prototype testing now with a planned launch in September 2025. The game engine is PlayCanvas (engine-only) and the platform is WebGL (Mac/PC/ChromeOS).
by archiepeach on 5/25/25, 10:13 PM
The poetry one is react native. Art and philosophy ones are swift/kotlin. I wanted to see if you could use LLMs to effectively create a cross-platform app. The idea behind react native was that you write it once in an approachable language, then the framework compiles to native app code. In 2025, the approachable language you code in is English, and the LLM now generates native app code.
It was generally a success and I feel less of a need of the development overhead of react native these days.
https://apps.apple.com/gb/app/for-arts-sake/id6744744230
https://apps.apple.com/gb/app/daily-philosophy/id6472272901
https://apps.apple.com/gb/app/the-poetry-corner/id1602552624
by ecralx on 5/27/25, 11:59 AM
The idea is simple:
1. On the days I work, I get an email one hour before the end of the day (as I usually use emails as todo lists).
2. I reply to that email with what I did just like I would in a personal journal.
3. My reply gets saved and shows up on a calendar. It's nothing fancy, just the content of the email on the day it was written.
4. When I need to prepare for a performance review or a manager check-in, I can generate a clean report based on my past entries with summaries written by AI
That’s it. No login every day, no new tool to open. Just a little trace of my work, saved quietly, and summarized when needed.
by DamnInteresting on 5/26/25, 1:49 AM
Now I need to figure out an alternate payment handler, or just give up on my modest monetization plan where players can pay a small one-time fee to unlock all of the archived puzzles. It was never going to make a fortune, but it would have been nice to offset some of the hosting expenses.
Don't use Stripe. They shit on you just because they can.
by aag on 5/25/25, 10:31 PM
by plankers on 5/25/25, 9:06 PM
now i'm starting on adjusting the model to include the liquid water ocean underneath the shell and observe the effect of changing viscosity gradients in the equilibration of the ocean and ice shell, as well as adding in compositional impurities (chloride brines) and tidal heating effects.
by hopeadoli on 5/25/25, 11:21 PM
An AI-trip planner with a nice twist. It shows you everything you need to know about a place even before getting there: Images, a great summary, cost of living broken down weather conditions etc. It also comes with the usual features you'll expect in a trip planning app (ai itinerary suggestions, travel expenses tracker, group chat for group trips, google places integration for looking up places to eat, things to do, healthcare places and transportation centers, and a private travel checklist). You should check it out today!
by czhu12 on 5/26/25, 12:39 AM
The two things I've open sourced that I've been moonlighting after hours for:
https://canine.sh/ - An open source Heroku alternative, 10x cheaper. Got sick of paying those prices
https://hellocsv.github.io/HelloCSV/ a free open source flatfile alternative
by carlnewton on 5/26/25, 6:06 AM
- The idea: https://carlnewton.github.io/posts/location-based-social-net...
- A build update and plan: https://carlnewton.github.io/posts/building-habitat/
- The repository: https://github.com/carlnewton/habitat
- The project board: https://github.com/users/carlnewton/projects/2
by mikhmha on 5/26/25, 12:51 AM
https://swarmmo.games The game is playable in browser! And theres no sign-up required to try it out.
by nullderef on 5/25/25, 8:32 PM
So I'm giving a try to a project which started with marketing. No implementation, just a TikTok to see if people like it. And holy crap, we got 75k views!
The new idea [2] is easier to explain (1 pushup = 1 minute of scrolling) and already has a community. Plus, not working alone helps me focus on what I'm good at: programming. I don't regret learning about other areas but doing marketing for a living is not my thing.
I'm not getting rid of SpeedBump, though. It's a fun side project and it does help people :)
by atlasunshrugged on 5/26/25, 5:37 AM
https://shorturl.at/LKMxS (Amazon link)
by gudzpoz on 5/25/25, 10:16 PM
[1] https://codeberg.org/gudzpoz/Juicemacs/src/branch/main/elisp
[2] https://www.graalvm.org/latest/graalvm-as-a-platform/languag...
[3] https://www.gnu.org/software/emacs/manual/html_node/elisp/Bu...
by bdxn on 5/25/25, 11:25 PM
A document specification for defining command line interfaces.
It's really just a fun side project to get more familiar with Go. The goal is to be able to generate boilerplate code in a few languages/frameworks and to generate documentation in a couple formats.
by Alex-Programs on 5/25/25, 10:24 PM
I've been doing a lot of experiments evaluating LLM translation performance, and I used what I learnt (that LLMs make mistakes, but different LLMs make different mistakes, and they're better at critiquing translations than producing them) to make a hybrid translator (https://nuenki.app/translator) that beats everything else.
And I was invited to do a talk about that to a company, which was really cool! I'm 19, doing this in my gap year before uni.
by popupeyecare on 5/26/25, 11:39 AM
It started because I wanted to send my sister an old photo for her birthday, and instead of using CVS or Shutterfly like a normal person, I ended up building it using Bubble and Lob.
Now I’m exploring the B2B angle — a friend asked if I could do postcard mailings for her medical practice. Curious if anyone here has experience with postcards for outreach or marketing?
by l72 on 5/25/25, 9:32 PM
Backend is already working: Boldaric https://github.com/line72/boldaric
And a simple iOS native front end (which I haven’t submitted to the App Store yet). Tor Jolan https://github.com/line72/torjolan
It has been interesting tweaking the algorithms and models for various similarity searches.
I really like that it focuses on music characteristics and not metadata, so popularity of a song/artist isn’t even taken into account. This has really helped me explore my rather large music collection especially when I get stuck in a rut of listening to the same things.
by jumski on 5/26/25, 10:07 AM
I have strong focus on db consistency and Type safety between steps.
### Built so far – core SQL orchestration (state machine, retries, observability, queue management) – strictly typed TypeScript DSL + compiler that turns flow definitions into migrations – a minimal Edge Function worker that polls, executes and reports back
### Current focus I’m heads-down on a dedicated client library. It leverages Supabase Realtime to stream events and progress of each flow run back to the browser, while having strict compile-time safety for flows defined with the TypeScript DSL.
### Next on the roadmap - Fanouts for processing arrays of data in parallel, with per-item retry logic. - SQL-based conditionals/branching that leverages JSONB containment operator and step outputs
Docs: https://www.pgflow.dev/concepts/how-pgflow-works/ Repo: https://github.com/pgflow-dev/pgflow
by piker on 5/25/25, 8:54 PM
Tritium is an IDE for corporate lawyers. Draft Word docs, review PDFs, redline all in a single application. It's written in Rust using a modified version of egui. Immediate mode has some interesting tradeoffs that I'd love to discuss on here. Also the web/desktop dichotomy presents a lot of interesting opportunities and challenges where data governance is concerned. I'd love your thoughts or to share mine!
by bibin765 on 5/26/25, 3:35 AM
It started as a side project to explore the latest AI trends. Now it’s something we use daily — and others are starting to as well.
Thoughtcatcher is a lightweight, AI-powered notes + reminders app that acts like a memory companion.
It helps you: - Capture raw thoughts and auto-tag them using AI - Set smart reminders triggered by context and meaning - This was a game changer for me personally - Search and chat with your notes like a conversation — not just by keywords, but by intent
Example? You’re walking out of a meeting and think: “We should revisit that pricing model after the new release.” You jot it into ThoughtCatcher — no structure, no stress. A week later, right before the next sprint planning, it reminds you. Just when you would’ve forgotten — it remembers.
What started as a learning project has grown into something useful — not just for individuals, but for teams too.
We’re now exploring B2B use cases like: • Project knowledge management • Shared team notes with smart search and chat • Meeting follow-up insights and reminders • AI-powered team memory for client or product work
Want to try it out? Android users: Download the app iOS users: Use the PWA — just “Add to Home Screen”
Still early. Still learning. But ThoughtCatcher already feels like something I wish I had years ago.
Would love your feedback or thoughts. And if you’re building something similar— let’s connect
by nicbou on 5/29/25, 2:51 PM
The system is complex. Choosing the best option is difficult. Understanding the timelines, the required documents, and how to prove coverage for a visa application or school enrolment is super complex. I'm trying to streamline the whole thing.
I released phase one yesterday. It's a simple overhaul of the text advice plus a new health insurance calculator. I also paired up with a health insurance broker to give more personalised advice, since everyone needs a different recommendation.
It's here: https://allaboutberlin.com/guides/german-health-insurance
I'm now working on phase two: covering the exact process of getting insured for a visa application. I really want to walk people through the exact order of things, with detailed steps and timelines. The calculator will get a hell of a lot more complicated.
Phase three is compressing all of that. Once I have a good idea of how things work and where friction can be reduced, I'll simplify and automate as much of it as possible. It should turn a messy, complex and confusing process into a fairly straightforward thing.
I feel like no one is really doing this yet, and I'm surprised at how achievable it is.
by runarberg on 5/25/25, 8:42 PM
What I have is a basic flash card app with double sided cards (for writing (i.e. drawing) the kanji, and reading). What sets it apart is that each card contains all the relevant dictionary data, and users are encourage to bookmark a couple of words to help them remember the writing or the reading of the kanji.
What I am working on now is the database backup/sync system. I store all the user’s progress in indexeddb store in their local browser. To sync I am writing a simple patch system, so they can pick a remote somewhere (e.g. a gist on github) and push their latest patches, when syncing progress I would check the hash of the patch and apply the relevant patches.
After that I am planning on turning it into a progressive web app so users can download the app onto their devises.
by ayugarg567 on 5/26/25, 4:32 AM
We used to run a content and product studio, and every time we had to get NDAs or service agreements signed, it felt ridiculous to pay $30/month for 2-3 documents.
So we built SignWith — no subscriptions, just simple pay-per-document pricing. ESIGN act and UETA Complaint.
Every month you get 3 free signatures - and then you can pay per document. So best ROI.
Works great for freelancers, early-stage founders, agency folks, or anyone who doesn’t need to send docs every day but still wants something secure and professional.
Launched recently. Early traction's been interesting. Would love feedback if you’re in this space or use e-sign tools regularly.
by CliffyA on 5/26/25, 1:18 AM
Currently it is being transpiled to C++ and C#
by crsn on 5/25/25, 10:45 PM
In particular:
To help solve forecasting & planning problems too hard to hold in your head, I’m converting natural-language formulations of constrained optimization problems into (back)solvable mathematical programs, whose candidate solutions are “scenarios” in a multi-dimensional “scenario landscape” that can be pivoted, filtered, or otherwise interrogated by an LLM equipped with analytical tools:
- 5 minute demo: https://youtu.be/-QdqiLp_9nY
- Details: https://spindle.ai
Eager to connect with anyone interested in similarly neurosymbolic “tools for thought”: carson@carsonkahn.com | +1 (303) 808-5874
by Ilasky on 5/26/25, 12:16 AM
It's an email-only Korean (and English) language learning service started by me and my partner! We've both been interested in language and, since I've been learning Korean and she's been improving English, I've been looking for a modality that works for me -- apps don't click for me for some reason.
It started by me sending myself an email everyday with new Korean vocab/grammar, but we thought it would be nice if it responded with corrections and learned from my common mistakes to make better questions. So, we've built it out to work for us, and turns out other people also like it and are growing everyday!
by ryanmentor on 5/26/25, 6:29 AM
Tech Pizza Mondays has been going well, as the Toronto HN meetup. If you're reading this on HN, you're welcome to join us -> URL Mondays.Pizza details on fedi.
We just finished Toronto's Pizza Day, our fifth year in a row and most successful so far. toronto.pizzadao.xyz will eventually get updated to next year's event invite.
I threw a fun experimental comedy party, and I plan to throw another in August. Having a volunteer cast was great, the clown burlesque was shockingly great. I hope to find some team members who want to join forces and put on the greatest talent show this town has ever seen!
Working on founding a school in Toronto has been very difficult, feels like I'm making zero real progress over time. If anyone cares deeply about education, I'd be happy to talk. I'm extremely enthusiastic about the Sudbury Valley School model.
And this last week I started a fitness club, where most weekdays we do strength training with giant wooden swords. Looking for people who want to join up and get strong and have fun!
by cmontella on 5/26/25, 4:09 PM
It's for building robots, scientific computations, user interfaces, reactive systems, and so forth. Currently working on docs for this new version that I'm about to release. I'll post a Show HN when I do. Here's the spec so far it's a very WIP, maybe 20% done: https://docs.mech-lang.org/v0.2/II.reference/specification.h...
The cool part about the spec is it's a valid Mech program which is formatted into html by the formatter! https://gitlab.com/mech-lang/docs/-/raw/main/II.reference/sp...
by tootyskooty on 5/26/25, 12:54 AM
A website for open-ended learning with LLMs. Something like a personalized, generative Wikipedia. Has generated courses, documents, quizzes and flashcards.
Each document links to more documents, which are all stored in a graph you grow over time. Currently using the graph for topic suggestions, though I've also been playing around with steering document generation with it as well.
by lihaoyi on 5/25/25, 10:04 PM
Build tools are generally an un-sexy field, and JVM build tools perhaps doubly so. But Mill demonstrates that with some thought put into the design and architecture, we can speed up JVM development workflows by 3-6x over traditional JVM tools like Maven or Gradle, and make it subjectively much easier to navigate in IDEs and extend with custom logic.
If you're passionate about developer experience and work on the JVM, I encourage you to give Mill a try!
by harisund1990 on 5/25/25, 10:20 PM
Think of it as a true drop in replacement for postgres that runs on multiple nodes. It internally does replication, sharding and leader election. Just add more nodes and you get to increase both read and write scale.
I personally am working on a few things like online major upgrades, async replication for DR, enhanced backup/restore/pitr/clone capabilities, and more recently supporting DocumentDB extension which provides a true Mongodb API.
Being a startup I also get to talk with large customers, help with marketing content, and participate in database conferences.
by frainfreeze on 5/25/25, 9:27 PM
Checkout how posthog did it [1], it's an interesting approach. Having something that can support both devs and content folks (non technical) is great. It is easy to get bogged down in building the website and reinventing bunch of wheeels, instead of focusing on the product & content, esp in smaller teams.
[1] How PostHog built a community forum, roadmap and changelog on Strapi https://strapi.io/user-stories/posthog
by arjunbajaj on 5/25/25, 11:22 PM
A developer-focused IoT Cloud Platform. The idea stems from pain points experienced while automating an indoor farm a few years ago where I had to spend way too much time building the data collection and analysis infrastructure instead of focusing on the actual automation.
Devices connect via secure MQTT, HTTP, or WebSockets and send structured, typed data. Each device gets its own sequential mailbox for messages. You can trigger webhooks or broadcast messages to other devices based on incoming data, powered by programmable actions.
Just deployed to production. Currently working on Device SDKs (coming very soon) and time-series analytics. Check out the platform, we're in technical preview now. Happy to answer questions and appreciate any feedback.
by bruno_rzn on 5/25/25, 9:29 PM
I have a software engineering background, and I’ve been working on this for nearly 3 years now! I used to play with the Electra One controller before, but having the encoders over the display is really something I’ve always wanted.
I presented Karl last month at Superbooth (a fair in Berlin) and got really good feedback. After 6 months of beta and 2 years of touring with it myself, the first batch will be dispatched in August, and this is quite exciting!
by jumploops on 5/25/25, 11:21 PM
Here’s a simple app my toddler made to generate toy trains[0].
“Real users” are using it to build personal software tools like finance dashboards, content generators, and educational apps.
Right now the functionality is great for many simple tools, but it’s notably lacking a first-class data layer (coming soon!).
All of the AI-generated code runs in secure MicroVMs, and the front-ends are just static assets, meaning the apps scale to zero when not in use.
We’re currently in the process of making the builder less of a “workflow” and more purely agentic, which should improve the overall success rate.
by whiskey-one on 5/26/25, 6:36 AM
The goal is for you to get complete and honest feedback from the team you work with. The killer feature is we make it quick and easy. And your reflectors can never initiate the instance of giving of feedback, the system invites them; this avoids the "but I am angry now" situation.
You can add the team members you work with in any setting and specify how often you work together. Depending on that, they will get the chance to reflect on you between every four weeks and every eight weeks. A reflection takes literally seconds.
Sign up is invite-only atm but invites are handed out pretty quickly since we want the feedback! Just sign up for the wait list on the site [1]. It will be free because we have a business version [2] that will support it; and the personal app will hopefully generate interest in the company one.
You can send feedback to "admin" at the domain of the company one [2].
[1] Personal: https://www.mymirror360.com/ [2] Company: https://www.mirror360.org/
by Seb-C on 5/26/25, 9:07 AM
by Cyphase on 5/26/25, 8:08 AM
Been a freelance dev for years, now going on "sabbatical" (love that word) very shortly.
Planning to do a lot of learning, self-improvement, and projects. Tech-related and not. Preparing for the next volume (not chapter) of life. Refactoring, if you like, among other things.
I'm excited.
---
I posted about this on last month's thread, and mentioned in a descendant comment that I was going to PyCon US in May. Someone well-meaning asked if tech conferences are in tune with working on myself. I just got back from PyCon. Yes! It was an incredible experience. Not for learning technical things from talks – the talks are mostly online already – but for the social and community aspects.
by marginalia_nu on 5/25/25, 10:10 PM
A bit of a janky setup, but I've mostly gotten it to do what I want it to do after some head scratching.
by Benjamin_Dobell on 5/25/25, 10:50 PM
For older kids I've been making it easier to write games in Godot using TypeScript:
https://breaka.club/blog/godots-most-powerful-scripting-lang...
I'm building tooling using this technology which allows kids to create their own games, this is itself presented as a game kids can play through. Basically, imagine if Roblox actually delivered on its promises to kids.
Most of what we're building will be open sourced, so that older kids / young adults will be able export their projects and share their creations stand-alone.
Of course, telling kids they can create their own game is only relevant is kids want to do that. We're not locked into one way of thinking. We've also modified Overcooked 2, a traditionally co-op game and introduced a visual scripting platform which allows kids to code their way through levels:
https://www.youtube.com/watch?v=ackD3G_D2Hc
Overcooked 2 won't be the only game for which we do this. Introducing coding to existing games is a fun way to teach kids to code, without yet burdening kids with too much creative freedom. Kids already want to play these games, so this approach allows us to bring educational tooling to kids rather than vice versa.
I used to be Head of Engineering at Ender, where we ran custom Minecraft servers for kids: https://joinender.com/ and prior to that I was Head of Engineering at Prequel / Beta Camp, where we ran courses that helped teenagers learn about entrepreneurship: https://www.beta.camp/. During peak COVID I also ran a social emotion development book subscription service with my wife, a primary school teacher.
by sidyapa on 5/26/25, 10:42 AM
https://chromewebstore.google.com/detail/hn-new-links-highli...
by zacharycohn on 5/25/25, 8:46 PM
A wordle-like game based on a road trip game my friends and I used to play. It serves you up a mashup of two different movie plots, and you have to guess the combined movie title. There's always some sort of shared word or wordplay between the two movie titles.
An example from the tutorial: the day after tomorrow never dies.
by taroth on 5/26/25, 5:50 AM
So I’m working on a universal progress bar HUD
- inspired by World of Warcraft raid mods
- fun sound effects for job start, end, error, and milestones
- can quick jump back to relevant app/tab
- starting with terminal commands and Claude code, cursor agent next
by vahid4m on 5/26/25, 1:06 AM
A one time payment (no subscription), local text to speech solution for MacOS or Windows. I've been a heavy user of Speechify but it was a bit expensive to justify. So I eventually built With Audio and after a few iterations now I'm working on the desktop app. Our laptops are very powerful and they are more capable of running some of these Text To Speech models with near to real time performance.
by Oras on 5/25/25, 8:27 PM
When I was looking for a job last summer, I got frustrated with the current resume builders on the market and decided to build one exactly how I wanted to use it.
- No signup, no login, and no payment.
- Suggest a professional summary (with highlighting) to match a job description [0].
- Preview as you go.
- ATS friendly templates.
- Find relevant jobs for my resume.
[0] Recruiters skim through resumes, and highlighting the keywords they look for has always helped me to get their attention, so I decided to implement this feature using AI.
by lancekey on 5/26/25, 2:21 AM
My thesis is that the next few decades will be driven by the prices of compute and kWh. This is my way of getting a better understanding of the ecosystem.
by psviderski on 5/26/25, 1:04 AM
Imagine Docker Compose and Tailscale had a baby. Uncloud is the baby that allows you to deploy and manage containers across a network of Docker hosts. I feel like we forgot how to build simple tools that do the job and not a thousand other unnecessary things.
I've started migrating some of my self-hosted web apps from my k8s cluster and I really enjoy it. Using the tool early on helps me better understand where the biggest pain points are and what I need to prioritise.
by jakevoytko on 5/26/25, 4:22 AM
In the past few months, I've learned that (a) writing a bulleted outline is a cheat code for producing decent work quickly, (b) every newsletter is a reminder that people could unsubscribe, so skip publishing issues that you're not proud of, (c) people really like stories, and (d) it's okay to have a mix of formats.
by berdon on 5/25/25, 11:03 PM
by murrion on 5/25/25, 8:31 PM
So I put together a simple Digital Asset Manager (DAM) where:
* Images are uploaded and vectorized using CLIP
* Vectors are stored in Lance format directly on Cloudflare R2
* Search is done via Lance, comparing natural language queries to image vectors
* The whole thing runs on Fly.io across three small FastAPI apps (upload, search, frontend)
No Postgres or Mongo. No AI, Just object storage and files.
You can try it here:
Or see the code here:
* https://github.com/gordonmurray/metabare.com
Would love feedback or ideas on where to take it next — I’m planning to add image tracking and store that usage data in Parquet or Iceberg on R2 as well.
by static_void on 5/25/25, 10:27 PM
Its real purpose is twofold: I enjoy data modeling, and doing just enough Rails work to regain fluency after a gap.
by jamesdhutton on 5/26/25, 2:01 AM
by satisfice on 5/25/25, 11:01 PM
This involves exploring the ethics of using magic to accomplish tasks. The problem then boils down to a matter of epistemology— a testing problem. But testing is something you only do in the absence of trust. So critical thinking begins with the rejection of trust.
It’s been interesting to read about “Anomalistic Psychology” which is the study of magical thinking. Malinowsky commented that not a single canoe was built by Melanesian islanders without the use of magic, yet none of them would say that they could be built without craftsmanship.
Magic is the belief in the infallibility of hope, to paraphrase Malinowski. Which may explain why too many smart people are uncritical about LLMs.
by unsoldbanana on 5/26/25, 1:35 AM
Here's a sample if you are wondering what its like: https://finzz.xyz/shared/rnl-oO955IPZN6Ux78E-l8DC
by mindcrime on 5/25/25, 8:19 PM
1. Right now, working on standing up an MCP server in Java. Not using the Spring Boot support at the moment, but rather setting up embedded Tomcat and doing it the more "low level" way just for didactic purposes. I'm sure I'll use Spring Boot once I get deeper into all of this.
2. Plowing through the "AI Agents in Action" book. I'm just wrapping up the section on AutoGen and about to move into crew.ai stuff.
3. Reading a book on Software Product Line Engineering.
4. I have an older project that's Grails based that I let linger without any attention for a really long time. I'm working on updating it to run on the latest Grails and Java versions and also writing some automated smoke tests.
by anonzzzies on 5/26/25, 4:15 AM
by yoav on 5/25/25, 9:42 PM
It’s an alternative to Electron/Tauri that uses Bun.
It has a bsdiff based update mechanism that lets you ship updates as small as 4KB, a custom zstd self extractor that makes your app bundle as small as 12MB, and more.
I’m currently working on adding Windows and Linux support.
by umvi on 5/25/25, 9:29 PM
This game was originally inspired by the game "Untrusted" (https://github.com/AlexNisnevich/untrusted)
by krlx on 5/26/25, 3:16 PM
Still quite early in the development but if you want to make my phone ring you can place an order on this page : https://micro-cafe.nuxt.dev/alix
You won't see your placed order nor get served however.
by dmitrysergeyev on 5/26/25, 4:41 PM
Goal: replace the vague boilerplate of listings with something closer to a paranoid friend who's done way too much research. Curious if this kind of subjective, context-heavy analysis adds real value—or just noise.
by StackRiff on 5/25/25, 10:48 PM
A social event planning app to capture the fun my friends and I had with facebook events, but without the facebook. We have native apps for iOS, Android and the web. dateit has a generous free features compared to competing apps (SMS invites, photo upload, customization).
My cofounder and I have fully bootstrapped this and now it mostly self sustains which is an exciting achievement!
It's been a fun project to hack on for the last couple years and spawned several interesting side quests. For example, the backend is in Swift (as I started as an iOS dev) so that has been an exciting space to work in.
by srcreigh on 5/25/25, 11:50 PM
I've added a react SSR system. It has node subprocess code for rendering HTML from java via stdin/stdout. There's a Node/Vite proxy server that adds the fancy HMR you expect from SPA apps.
It supports multiple roots on a page, every SSR component has data-props and data-componentname, and the entry script just queries those attributes and hydrates everything.
The node renderer script is packaged as an EXE which is deployed in WEB-INF on the server.
It's fun to add the amazing React tooling to an old codebase. It also shows how you really, really, really do not need NextJS.
by straumat on 5/26/25, 6:26 AM
To tap into this revolution, I created Mogami: a Java / Spring Boot library that lets you make any URL pay-to-access with a single annotation. → Code & docs: https://github.com/mogami-tech/x402-spring-boot-starter
by wagslane on 5/26/25, 2:09 PM
Like we give people challenges for each new concept, but we don't give them 4 or 5 challenges for each new concept. It's hard to get right because
1. It's a lot of work to make challenges/review material 2. Every student wants a different amount of practice/review
So anyways I think we've got a really good solution in the works, been cooking on it this week and we're hoping to be ready to go live in August
by admiralrohan on 5/26/25, 11:29 AM
It stems from my struggle to keep up with the feeling of overwhelm with so many things to learn. Currently working on organizing all of my thoughts into a coherent theory.
You can check my blog to learn more about my journey and the theory = https://www.moderncynicism.com/about
by flippyhead on 5/26/25, 1:03 PM
I got sick of starting a project only to later discover significant competitors or features of the marketplace I missed, despite looking. I can now do in minutes what used to take me days.
by qu0b on 5/25/25, 9:19 PM
by mukeshsoni on 5/26/25, 7:34 AM
I have built the basic photo management functionality. I have also added conversational search using the CLIP model, which is working really well.
I hope to add a photo editor to the app because that's the part i love about Lightroom. It's a photo management app with a built in photo editor (which is really good).
by matty22 on 5/26/25, 4:47 PM
by gue-ni on 5/26/25, 6:53 AM
by stuartdev on 5/26/25, 5:58 AM
- https://bookreadinghabit.com/ - https://apps.apple.com/app/book-reading-habit/id6742913266
by ecce_homo on 5/25/25, 8:40 PM
by monroewalker on 5/25/25, 10:52 PM
It’s just a basic IntelliJ plugin which provides an infinite canvas to add code bookmarks to. I work on a large code base and often have to take on tasks involving lots of unfamiliar areas of code and components which influence each other only through long chains of indirection. Having a visual space to lay things out, draw connections, and quickly jump back into the code has been really helpful
The canvas and UI is built using Java AWT since that’s what IntelliJ plugins are built on, but it occurred to me that I could just throw in a web view and use any of the existing JS libraries for working on an infinite canvas. React Flow has seemed like the best option with tldraw being what I’d fallback to.
But then.. if the canvas is built with web technology then there’s no reason to keep it just within an IntelliJ plugin vs just a standalone web app with the ability to contain generic content that might open files in IntelliJ or any other editor. I’m pretty sure the “knowledge database on a canvas” thing has been done a number of times already so I want to also see if there are existing open source projects that it’d be easy enough to just add a special node type to
by colordrops on 5/26/25, 5:51 AM
A turnkey self-hosted home router + app server that you can (eventually) just plug into your modem and go through a simple setup and have self hosting. It's in heavy development, so it requires some experience with NixOS to get running, but the goal is to have it usable by non-technical people.
Current features
- Firewall
- Headscale VPN, apps private by default
- Automated DDNS
- Dozens of apps with automated subdomain allocation and TLS cert management
- Ad Blocking
- Customizable public landing page
- Single module configuration for all features
- Automatic backup for all apps
- Better landing page and contribution process
- Admin Dashboard, Config page
- SSO across apps
- Ability sync to a cloud instance in case you no longer want to host at home
- Ability to mesh with friends/family/associates for social networking, distributed backup, distributed CDN, etc
- Getting others to install it and build a community
- A standard protocol or tool for updating DNS entries at registrars
- A standard protocol or tool for updating modems with suitable settings for self-hosting
- a free DDNS service to provide to users of HomeFreeby seveibar on 5/25/25, 11:25 PM
by max_on_hn on 5/26/25, 7:18 AM
It works by connecting directly to Linear and dispatching assigned tasks to agents that submit PRs in GitHub when finished. My agents work in a fully-integrated Linux development environment, including internet access. This means that they can browse the web, install dependencies, and creatively work around environment issues to make sure they run and test the code they ship.
It's really gratifying to see people asking all over the internet, "Where can I just create tickets and get pull requests?" because that's exactly the workflow I built CheepCode to support. As an engineer for almost 15 years, I knew what I personally wanted, and it really makes me happy to see that what I built will work for so many others too.
As a bootstrapped solo founder, it's challenging to juggle product/growth/development/strategy all at once, but also incredibly rewarding. I wouldn't necessarily say no to funding ;) but in the meantime, it's quite a thrill!
by Daenks on 5/26/25, 1:42 PM
by wolfoftheweb on 5/26/25, 3:05 AM
It uses GPT-4 for story generation, ElevenLabs for narration, and a simple Next.js + Supabase stack for the app layer. I’m experimenting with story memory (so kids can revisit recurring characters) using vector embeddings, and building a “choose-your-own-adventure” mode with dynamic audio rendering.
Biggest challenge so far: aligning narration, ambient sounds, and story pacing without sounding janky or robotic. Solved it by tokenizing and chunking the story for synchronized audio stitching via ffmpeg.
Another challenge was the inconsistent image illustrations via Dall-E 3. I’ve adopted a dynamic prompting method that includes as many details about the scene, character details, and other visual elements which should remain consistent on each of the storybook pages.
If your kid ever demands “one more story” after a long day, I built this for you. It’s had a meaningful impact on my son’s behavior.
by petabyt on 5/25/25, 10:59 PM
by fmstephe on 5/26/25, 1:34 AM
In contrast to the popular arena based allocators (which target quickly allocating/freeing short lived per-request allocations), I am targeting an allocator for build very large in-memory dbs or caches with almost no garbage collection cost.
There's a little no-gc string interner package in there as well.
https://github.com/fmstephe/memorymanager
It's somewhat on pause right now as I have just started a new job. (but it has been a very fun project, nerdy joy).
Related to the memorymanager, as in intending to support it are
https://github.com/fmstephe/fuzzhelper A library for setting up fuzz tests for complex data structures.
https://github.com/fmstephe/gossert A library for adding runtime assertions to Go code. It's developed so that when the assertions are switched off the compiler should be able to completely eliminate the assertions. But this requires build tags to switch the assertions on.
by parsabg on 5/25/25, 11:20 PM
You can use it to check and summarize news and social media, fill out forms, send messages, book holidays, do your online shopping, conduct research, and pretty much anything else that can be done within a browser.
by jrslv on 5/26/25, 2:53 PM
It aggregates job ads from various sources and lets employers post directly. The goal is to help more people find meaningful work in wind, solar, BESS, etc.
Built with PHP, PostGIS, and a swarm of GPT-assisted cron jobs. I’d love your feedback.
by NiloCK on 5/26/25, 11:59 AM
I want to sand away the roughest edges of the SRS user experience, and to enable individuals and organizations to easily spin up courseware that's aligned with best pedagogical practices and fits naturally into the hands of educators.
And, in the current wave, also think bolt.new or lovable.dev, with 'agentic' workflows for both content (scouring source material for ) and for user-authored bespoke interactive, via LLM authored artifacts.
Optimistic for some ShowHNs or more general shilling this week.
- https://github.com/patched-network/vue-skuilder - https://patched.network
by synergy20 on 5/25/25, 11:29 PM
My middle school aged kids was able to escape with free proxy, vpn, tor etc in the past which forced me to figure out a way to lock it down totally when it's absolutely needed.
by dhamidi on 5/26/25, 8:58 AM
Growing a coding agent as a "game": https://github.com/dhamidi/smolcode. Started with https://ampcode.com/how-to-build-an-agent and then asked the agent to modify itself. It's so much fun and hard to stop!
Since coding is cheap now, I'm building a web framework for myself, using reasonable defaults: built around the transaction log, single binary deploys, sqlite3 for everything, runtime introspection tools to make ops easier.
I want the framework to "compile itself out of the project", so that projects don't have devtime dependencies and can be easily edited by Cursor et al. It should still stick around in the form of a high-level CLI that an AI agent can leverage to fulfill common tasks (e.g. add a new UI component, query, command, etc).
by brylie on 5/26/25, 11:59 AM
My co-founders and I are building an AI assistant that helps immigrants navigate Finnish bureaucracy. As immigrants, we've experienced firsthand how fragmented and inaccessible essential information can be, scattered across Migri, Kela, the tax office, and municipal websites, often only in Finnish and Swedish.
We're using RAG (Retrieval Augmented Generation) to create a multilingual chat interface that connects people to official resources rather than replacing them. The system reduces "failure demand" - support requests that arise because people can't achieve their goals through existing resources.
The technical approach is a multi-agent system in which specialized AI agents handle different domains (immigration law, employment, housing, healthcare, etc.). We've named the agents after Finnish mythological figures (such as Tapio, Ilmarinen, and Sampo) to create a cultural connection while providing practical assistance.
Interestingly, this addresses a systematic problem - government agencies spend significant resources on repetitive support that could be automated. At the same time, immigrants get frustrated trying to piece together information from multiple sources.
We're exploring B2B opportunities (companies relocating employees, municipalities, healthcare systems recruiting internationally) and EU funding for integration technology.
Happy to share more details or get feedback from anyone who's worked on similar multilingual AI systems or government-facing tools.
by vivzkestrel on 5/26/25, 5:54 AM
- For node, that means it installs eslint + prettier / biome,
- setups all the plugins,
- sets up typescript with watch mode using tsx and alias support,
- installs vitest with all configuration,
- installs dotenvx and configures it to use separate development, staging, test and production configuration,
- installs commitlint, commitizen to ensure all your commit messages adhere to specific conventions,
- installs lefthook / husky to run tasks before commit, on push etc.
- Adds github actions for linting, running tests.
- Creates a docker development, testing, staging and production setup with different dockerfiles for each.
- Development version installs self signed localhost SSL certificates for use by nginx/traefik/caddy.
- Test environment runs all tests inside the container and shuts it down,
- Staging environment mimicks real production as closely as possible with minimal resource consumption.
- Production environment uses actual SSL certificates issued by the likes of zeroSSL or letsencrypt with a highly optimized dockerfile for minimal footage.
- Direct deployment to AWS / Azure / GCP
- Python version does a tonne of stuff too. Will share if anyone is curious
- What is special about this? As dependencies change, the boilerplate also updates itself, runs the tests and passing configuration is cached
by paulnovacovici on 5/26/25, 12:06 AM
A way to store bookmarks all in one place similar to Pocket, but built around semantic feature as the primary feature. Been beta testing an iOS app, but need to pivot on the name since there’s another Recallify on the App Store, and haven’t gotten around to it due to not much user growth.
by ctbellmar on 5/26/25, 12:05 PM
https://github.com/bikemazzell/whatsignal
I'm working on a WhatsApp to Signal relay. I.e.: whenever someone sends a WA message to you, it appears in your Signal. You can reply and it will go back to the original sender.
Why? I'm privacy conscious and don't fancy using a Meta product. But some of my friends/associates/family still insist on WhatsApp only. Running this WhatBridge service on my micro server behind a VPN allows me to communicate without having WhatsApp on my mobile.
Behind the scenes, it connects WAHA (https://github.com/devlikeapro/waha ) and Signal CLI (https://github.com/AsamK/signal-cli). Still early stages, but getting closer to a workable state.
by alfonmga on 5/26/25, 6:32 PM
yt2brave[0], a MacOS app that automatically opens YouTube links in Brave Browser for a better browsing experience on YouTube (no ads). It's compatible for Safari, Google Chrome and Edge. It is magic.
by Dathuil on 5/26/25, 9:06 AM
End goal is to have many reference examples when I start to make the game I want to make.
by tallytarik on 5/25/25, 9:40 PM
I've found it really satisfying to solve the data challenges that come along the way, from "where on earth could this data come from" to collecting, storing, parsing, validating and serving constantly. It's also - by nature - something that's never going to be "done". There's always something to improve. I love it!
We now offer more types of data (ASN/whois, proxy/threat detection, so on) than most other providers, more accurate and more frequently updated, at a tenth of the cost, which is something I'm really happy about.
For anyone interested, you can make 1,000 requests day free, or reach out if you have an open source/public interest project for an unlimited key or access to the data.
I'd also love to hear any suggestions for additional data types to add.
by trikko on 5/25/25, 10:08 PM
Serverino is a small, fast, and dependency-free HTTP server implemented in D. A minimal app with serverino can handle on my laptop ~150k reqs/s and it uses just a few mb of ram.
by shayanbahal on 5/25/25, 8:29 PM
- https://padsnap.app/ : PadSnap is a simple web app that adds customizable padding to your images so they fit Instagram’s/custom dimensions — no cropping, no quality loss. All on browser, no server uploads. Also no ads or login.
- https://shiryakhat.net/ : redid my podcasts website last week: Shir Ya Khat podcast, which translates to "Head or Tails" in Farsi, began its non-profit journey in 2016 with a mission to make blockchain and cryptocurrency technical knowledge accessible to Farsi speakers worldwide.
- life timetime visualizer, still WIP, feedback welcome: https://shayanb.github.io/timeline/
by jaronilan on 5/25/25, 9:33 PM
https://github.com/jaronilan/stories/blob/main/Base%20Rate.p...
Will now move at the usual snail pace to write the next one.
by henko on 5/26/25, 11:31 AM
by theThree on 5/25/25, 8:18 PM
by user432678 on 5/26/25, 1:03 PM
by klaaz0r on 5/26/25, 5:17 AM
We monitor how companies/brands rank in LLMs/AI Search, we don't use the API but the user interfaces (since these are totally different). We run thousands of prompts to analyse responses and combine that with website traffic to get an understanding of what is being said and how well you rank.
I didn't anticipate Google moving to AI mode so fast, to be totally honest, which makes it really interesting because we likely see rank tracking and keyword tracking disappear in the near future.
website: https://promptwatch.com
As a fun side project, I run https://homestra.com a kind of Zillow for Europe but for second/vacation homes
by oulipo on 5/25/25, 8:13 PM
by d4mi3n on 5/25/25, 11:16 PM
https://github.com/damien/content-security-policy/tree/main/...
Once I’m happy with my take on a reference implementation I’m hoping to create some tooling with it to do some interesting analysis of CSP abstract syntax trees to identify things like policy anti patterns, reporting on capabilities a policy grants to a domain/resource, and a better mechanism for allowing tools like OPA, SemGrep, etc. to define and enforce rules on a policy.
by linsomniac on 5/26/25, 12:45 AM
https://github.com/linsomniac/lessencrypt
Short description:
- Client generates RSA keypair and connects to server from <1024 TCP port, sends pubkey. - Server uses reverse DNS to come up with cert name (rules can specify alternate CN and SANs, override TTL, etc). - Server generates a signed cert and connects back to client on <1024 TCP port and sends cert.
by ginkgotree on 5/25/25, 9:47 PM
by csjh on 5/26/25, 4:31 PM
Also: the README is pretty outdated, the real interesting stuff happens in https://github.com/csjh/mitey-jit/blob/main/backend/arm64/ba...
by euvin on 5/25/25, 8:23 PM
1) a note-taking workflow in Obsidian (you take bite-sized notes about a topic, then connect "prerequisite" notes in Obsidian's canvas editor)
2) a tool that uploads each note and graph data to a database
3) a webapp that presents those notes algorithmically using spaced repetition. This enables you to allow others to "traverse" your note graph in a guided and self-paced manner.
You can add "challenge presets" to each note so that your mastery of each piece of knowledge can be tested with simple flashcards, multiple choice, free response, or some visual/actionable task to force active recall. An algorithm uses your success rate and spaced repetition data to introduce & drill more advanced notes into your long term memory.
Here's some more reading I was inspired by:
https://www.mathacademy.com/pedagogy
https://www.justinmath.com/individualized-spaced-repetition-...
Even if there are a lot of imperfections and flaws about this project (like the sheer difficulty of curating a good knowledge graph to begin with), I'm hoping to make my note-taking in Obsidian more structured and thorough, replace my Anki routine, and make any of my notes into an automated + algorithmic course. If someone has another similar project (combining note-taking with hierarchal, topological knowledge graphs with spaced repetition and testing all in one platform) I would love to hear more about your approaches. Quick shoutout to one person I've seen who is doing something similar: https://x.com/JeffreyBiles/status/1926639544666816774
by wateringcan on 5/25/25, 11:06 PM
A tool that scans California legislation and flags bills that might affect your startup. You drop in a link to your website or a short description, and it returns plain-English summaries and impact analysis of relevant bills.
by thephyber on 5/25/25, 10:58 PM
Separately, working on building an app to assist with cipher analysis of things like Kryptos and Bitcoin/crypto puzzles. Loosely modeled after CyberChef, but a native app that is capable of far more detailed frequency analysis and brute forcing with the GPU.
Also, experimenting with LLM workflows for both work and the rest of life. Prompt engineering seems like an incredibly valuable skillset for the next decade.
by clone1018 on 5/25/25, 9:05 PM
Really hoping to get some early feedback on this tool, I've been using it for two production sites for about a week now and I've already discovered (at work) that we've had the 2nd largest user signup day, and that we deployed a change that inaccurately tracked a specific metric. Check it out at https://querycanary.com
by ryansworks on 5/25/25, 9:37 PM
by baduiux on 6/1/25, 9:54 PM
I always felt like collecting all my todos / notes in a single notes app is not enough. Sometimes you want to remember something just for the next time you open this app/window. Attaching the note directly onto the window makes sure you won’t forget on it.
by zzsshh on 5/26/25, 11:09 AM
- Works by selecting traits, which then trains a model(flux) OR uploading your own photos.
- Hopefully useful for shopify/ecomm store owners, digital agencies, or just about anyone to drastically speed up the time and reduce the cost to test creatives, create media content, or just do what you want with a model of yourself.
- Web with Nextjs, DB with Supabase, marketing with nothing.
- Happy to share more details into how we built this, and hear feedback and thoughts.
by mindwork on 5/28/25, 12:34 AM
In short: installed immich(instead of Synology Photos), Beszel, Paperless-ngx(configuration pending), moved a bunch of tools out of my HTPC VM in to separate VM, did a proper backup of my proxmox cluster and set up cloud-init template with new ubuntu.
Due: gitops style deployment and terraformation of infrastructure(provision proxmox VMs via terraform) and auto redeployment of docker containers across 5 machines. Probably would need github runner in my network. Also Authelia is still missing in my lab, I need to get to that soon
by bryanhogan on 5/25/25, 11:24 PM
App will be local-first and without locking important features behind a subscription.
Very recently I finished my bachelor thesis which was about this app (focus usability and market fit).
Also made this site a few days ago, get notified when it launches: https://dailyselftrack.com
More about me here: https://bryanhogan.com
by yu3zhou4 on 5/26/25, 7:53 AM
by harundu on 5/26/25, 4:02 PM
Instead of integrating, maintaining and paying for multiple tools, everything is available in one simple to use dashboard.
For those familiar with PostHog, you can think of it a simple version of that.
I've been developing it and testing with few startups and plan to start promoting it more soon once the MVP is bit more polished. Also more tools are on the roadmap.
by sampullman on 5/26/25, 3:34 AM
I made it because the desktop software I use (ImageOptim) doesn't support Jpegli, which I find to be much better than mozjpeg for most types of jpegs used on the web.
There's still a lot of missing pieces, like adding color quantization and other PNG options, improving the UI, parallel webworker support, etc. The code is open source and can be used as a library: https://github.com/sampullman/image-opt
by yonatan8070 on 5/26/25, 6:10 AM
I started the project around a year ago, then got a little stuck and forgot about it for a while, until now.
For anyone curious about the technical details, it's a Rust program that wraps `ssh`, copies a small statically compiled binary to the remote, and uses SSHFS/SFTP (haven't decided yet, maybe configurable?) to allow your local editor to access the remote files as easily as local ones
by omgmajk on 5/26/25, 11:25 AM
by AlbinoDrought on 5/25/25, 8:20 PM
In the past few months, I've finally started working on a basic replacement NVR that works for me: https://github.com/AlbinoDrought/creamy-nvr
Like many video projects, it's a glorified ffmpeg wrapper :)
by mattbettinson on 5/25/25, 8:46 PM
by curo on 5/26/25, 2:26 AM
Until recently, it was cost-prohibitive to gamify topics with indefinite answers and progressions. As a result, "left-brained" topics have been gamified for years, but "right-brained" topics have resisted gamification. LLMs and generative AI unlock game economics for unstructured text.
Casual gaming is only category outpacing passive, social media. The most direct way to elevate humanity's media appetite is to turn human greatness into casual games.
by bennydog224 on 5/26/25, 9:46 PM
We currently are focused on hydroelectric power in the PNW region - converting river flow to tradable metrics.
Check us out -> https://askcrystal.info/dashboard
by maz1b on 5/26/25, 12:07 PM
We're looking on scaling further internationally. I myself am a US-Pakistani, the first medical doctor + full stack engineer and technologist in Pakistan's history (~250m people). Would love to chat with anyone interested!
by prashantsengar on 5/29/25, 1:11 PM
The idea came since Omnivore decided to shut down. It is open source and self-hostable but the whole system has so many bells and whistles that nobody would want to self-host it. Even more, even if I am able to export all my data, I will be unable to use their beautiful client to view my data or build anything of my own because I am not good at that.
The idea is to have a read-it-later app that has modular components each of which can be self-hosted independently which talk in protocols. So a different data crawler module that can take many forms - browser extension, Telegram bot, or something else, that takes an URL and posts the page content to a server.
The server will have modular parsers that are extensible to support different websites and pages, contributed by community (similar to Telegram's parser or Calibre's). It then converts it into a common format to be stored.
The storage should also be pluggable so that I can easily store my data wherever I want - whether locally as MD files, on S3, or on an SQL Database. This also should be extensible to allow different storage types like so many other tools do.
Then a web server that serves common endpoints to read, add tags and annotations, and other operations. It just needs the user's storage endpoint and authorization to allow them to access their data. I feel this is going to be the most challenging.
Let users create their own clients that render this for users. Just like Telegram and Matrix.
This idea resurfaced recently because I wanted to try getrecall.ai but it looks too expensive and I am worried about privacy. Having a setup like this would allow me to plugin a local LLM to this system as well.
by xandrius on 5/25/25, 9:15 PM
Planning to have a first testing session some time next month. Really excited but still lots to go!
by jason_zig on 5/26/25, 11:54 AM
I'm now supporting over 30K clients and over 40 million survey responses. Naturally lots of things come up when scaling a project solo heres a few:
- Optimizing existing reporting dashboards
- Improving onboarding experience
- Tapping new growth channels (Organic SEO, Paid ads, Integration marketplaces)
- Resolving customer support tickets and minimizing ticket flow in general
Lots of things come up which always keeps the work interesting. It's probably time to scale past one person though so that's next on the docket!
by smeej on 5/25/25, 9:56 PM
All I really want to do is be able to clip/save articles (and maybe generate transcripts from videos) from my phone or computer, read them in KOReader on a Boox tablet, and then export them and my eBook notes into Logseq, but every time I think I have it figured out, some project pulls a rug out from under me and I end up back at the drawing board.
by axegon_ on 5/25/25, 7:45 PM
by 0x000xca0xfe on 5/25/25, 9:49 PM
Also another fun idea I want to try is to let the Claude design a new programming language, i.e. where the AI makes all the decisions and goal-settings and I just help it instead when it's stuck.
by jlivin on 5/26/25, 3:24 PM
by ludee0 on 5/26/25, 8:12 AM
by azianmike on 5/26/25, 3:56 AM
Andrew Wilkinson tweeted the other day "I just found out how much we pay for DocuSign and my jaw dropped." https://x.com/awilkinson/status/1892638803505868824
So I thought "how hard could this be?" and decided to build a free alternative that's UETA and ESIGN act compliant!
by kenrick95 on 5/26/25, 12:36 PM
This is an example view-only trip: https://ikuyo.kenrick95.org/trip/2617cd98-a229-45d4-9617-526...
So far it has:
- Timetable view, with drag-and-drop support. I think it will be useful for planning the activities as I need to be able to quickly swap activities - List view. I think it will be useful for actual travel itself as the day go, I'll need to focus day-by-day - Maps. To make it easier to plan nearby activities, but I think I still need to rely on Google Maps or other kind of maps since I still need the navigation/direction in-between activities - Sharing & comments. Since I plan to use it to collaborate the travel with my friend, this is a must-have feature for me. - Expenses. This is still quite basic but maybe I should also track who owes whom...
I think now I'm happy with the feature set and will start planning my travel using this to polish the UX :)
by pizzly on 5/25/25, 11:04 PM
Currently working on making it even more reliable, navigate pages and understand website images not just the text of the webpage. Also getting ready for a Product Hunt launch.
by pseudocomposer on 5/26/25, 12:47 PM
Once I’ve got at least the iOS and web release pipelines working again, I plan to replace its Reddit integration/sharing features with an implementation based on https://jonline.io (https://github.com/JonLatane/jonline), the AGPL, Rust/gRPC-based social network I’ve built in the intervening time.
I’m also planning on adding more detailed (left to right audio) panning and other output controls, oriented towards using BeatScratch live with other musicians. My hope for this is to use it to complement my own musical practice I’ve been working on, outside of all this 1s-and-0s programming time.
by puskark4 on 5/27/25, 3:58 PM
Built the first product, which is email based where users email/forward suspicious messages to check@isthisspam.org and get a response back. Next up is a chrome extension, for now will have to use cloud providers but hoping to make it fully locally
by quantas on 5/26/25, 9:31 AM
by TkTech on 5/26/25, 12:55 AM
https://github.com/tktech/chancy
Chancy is a postgres-backed task queue for Python that scratches a lot of the Celery itches. It's not meant for folks that need to run a million tasks a second, but for the majority of projects (many millions per day) it offers:
- Robust job support, including timeouts, memory limits, retries, global uniqueness, global rate limiting, scheduling (cron and "in 10 seconds"), priorities, etc...
- DAG based workflow plugin
- It's asyncio-first with support for threading, multiprocessing, asyncio, and sub-interpreter tasks so each job can use the optimal concurrency model. Workflows can mix tasks across multiple queues and concurrency models.
- Can be embedded inside your existing ASGI servers - great for things like development docker containers or containers deployed on say, unraid.
- Worker's handle scheduling (no need for `celery beat`) and have an optional built-in dashboard.
- 1 infrastructure dependency (postgres) and 1 required package (psycopg3) - with everything else isolated in optional plugins. - Dynamically re-assign queues to new workers based on tags, add, pause (with auto-resume), modify or destroy queues at any time.
- Highly observable - unlike Celery, you can just query your database when needed to see the entire system state
- Portable - Linux/Windows/OSX
- Permanently free and open-source without any "premium" or paid features.
- Django integration - ORM/models, admin, and django auth integration for the dashboard.
Chancy is a young tool, but is used in production environments with tens of thousands of users and billions of jobs run with great feedback from early users:
> ...thank you for this amazing library...
> hey, first thanks for such a great library! Chancy has worked incredibly well, and its modular design has made it a pleasure to use. Super lightweight but feature-full; it's a hard balance to strike.
by tunesmith on 5/25/25, 8:35 PM
Along the lines of predictionbook, metaculus - something that helps you be "well calibrated", but more playful/fun than metaculus.
It doesn't have a lot of upside - predictionbook actually went offline due to lack of interest. But it was a good excuse to try out some vibe coding, and learn react native (I've mostly been a backend programmer).
In an attempt to make it more engaging and fun, I decided to have it focus on sports picks. Also partly because calibration graphs need to have a lot of predictions to yield any reliable information about your calibration.
I got it up in time for March Madness and about 25 of my friends joined and it was a good time. I nagged and reminded them a lot about about 15-20 of them predicted all 63 games, by picking the winner of each match and what their percentage confidence was. I had a leaderboard and live-blogged and gave silly awards.
I later added support for multiple "tournaments" and currently have tournaments going for NBA Playoffs and NHL Playoffs, but interest is waning. Of my friends, only 2-3 others are still regularly predicting.
Maybe it'll be more fun for the NFL season but I might also let it go a bit dormant.
Biggest challenge is that there isn't really a bulletproof way to rank people if people only predict some games in a tournament. I've tried all sorts of things, minimum # of games, bayesian kernel smoothing, but it's ultimately arbitrary when choosing how to penalize someone for not participating.
If I were to continue I'd be looking at things like automatically integrating with sports apis and odds/bookmaking apis, allowing users to create their own tournaments, etc. But ultimately, the UX of the site isn't much more than making a prediction, and then checking back later when the game is over to see your score. Not much more reason to hang around on the site than that.
by BrandiATMuhkuh on 5/26/25, 2:10 AM
Now, with LLMs, this got much simpler. As LLMs are really good at mimicking humans.
The first application will be to simulate a LinkedIn post against your own (virtual) followers. I grab your actual LinkedIn followers, "turn" each one into an LLM and see how they react.
by serezeapp on 5/29/25, 2:53 AM
Sereze guides personalized breathing techniques like Box Breathing, Yoga Nidra/NSDR, and Silent Meditation, tailored using your Apple Health data (HRV, HR, sleep). It works completely offline with no logins or server sync — just phone and your Apple Watch.
It tracks heart rhythm and variability in real time during sessions, then uses LLM/GenAI to offer personalized insights into your nervous system health, sleep recovery, and stress resilience. You can even ask it questions post-session and receive contextual responses.
Recent features include TTS-guided breath cues, REM/Deep/Core sleep stage breakdowns, and a silent mode for mindful journaling or meditative tracking.
Would love feedback or collaborations! https://apps.apple.com/us/app/sereze/id6736566597
by I_am_tiberius on 5/26/25, 3:17 AM
by denys_potapov on 5/26/25, 8:06 AM
Right now, I'm building a Space Invaders clone in Elixir with LiveView, and integrating Blockly so the game's core logic can be edited visually. Hoping it becomes a fun way to learn both functional programming and web dev.
by monkaiju on 5/25/25, 10:42 PM
A community powered, wikipedia-like, database for tracking police and their activities.
by bilater on 5/27/25, 7:55 AM
I've been experimenting with a custom workflow to turn long-form videos into short, engaging highlight clips. I know a bunch of startups are in this space, but I wanted to build something lightweight and flexible just for fun — and ended up integrating it into one of my products.
Here’s the general flow:
Transcribe the video Use tools like Whisper, AssemblyAI, or Deepgram to get an accurate transcript.
Extract interesting clips Feed the transcript into an LLM (I recommend Gemini because of its long context and quality) and prompt it to segment the video based on criteria like "viral", "engaging", "funny", etc. Make sure it returns timestamps.
Generate the clips Use ffmpeg to slice the original video using those timestamps.
(Optional) Auto-crop for vertical If the user selects a mobile/short-form format, use something like Sieve to auto-crop and center the subject for vertical (9:16) output.
(Bonus) Enhance captions Run the extracted clip transcripts back through the LLM to pick out keywords or phrases to emphasize in the captions.
Add captions with styling Use Remotion or similar to render the clips with styled, animated captions. The component handles logic for timing and highlighting.
Render and download Batch render your clips and you’re done.
I used this exact pipeline to build a feature in one of my tools: https://www.shortsgenerator.com/highlights-generator
Appreciate any feedback!
by manuel2258 on 5/26/25, 7:27 AM
by RobinL on 5/25/25, 9:15 PM
Unless I'm missing something, it's amazing how few free, _high quality_ materials are online.
Ultimately I'm interested in two things: genuinely fun games that make you do some maths, and quality visualisations that help make concepts easier to learn
by jobswithgptcom on 5/25/25, 10:17 PM
by caseysoftware on 5/26/25, 1:44 AM
by dvcrn on 5/26/25, 3:40 AM
It started out as a cloud runner for small (mostly pure) composable javascript functions that we run for you in the cloud. So whenever you have some small code snippet that you want to run, you can hit up microfn.dev, paste your code and then plug it into wherever you want (pipe into the terminal, use from MCP, add a cron to it, add to Siri shortcuts, use for home automation, ping it with webhooks, etc)
Now we added agents support, so you can have autonomous AI agents take your functions and decide when to use them. Even better, you can hook microfn into a MCP and have anything that supports MCP use those functions as well!
Suppose you need a new thing that your agent should do, you could ask the agent (whether that's claude or cursor) to compose a new function, add it to microfn, then use it itself going forward!
microfn.dev wants to be a toolbox for composable small tools. Imagine a toolbox at home with a bunch of hammers and screw drivers that you collect, share, and use for different purposes.
Some actual examples:
- Pull data off services (twitter, etc)
- Store data from sensors somewhere
- Give agents tools to talk to your specific systems without needing to write an entire MCP
- Wrap complex logic (eg slack auth + sending a message to slack) into a function and add it to a Siri shortcut, so you can quickly send messages to a specific channel with a ping or curl
Still very early alpha (beta-ish), but very excited about this
by Thomas_perez13 on 5/29/25, 6:28 PM
Hey everyone, I’m building my next project and would really value your input. I’m exploring two directions — both designed for mid-to-senior technical builders: AI Agent Builder: Create complex, production-ready agents from plain text in minutes. Fully code-ownable, transparent (not a black box), and easily connectable to modern tools — even the latest YC startups with APIs. Cursor for APIs: A dev-first tool to connect to any API instantly. Just type “build a RAG system for…” and it suggests the best tools, then generates the right code and surfaces the latest docs — including niche APIs. Think of it as a fast, intelligent API library with copy-paste-ready code. Which of these would actually improve your workflow?
by anttiharju on 5/25/25, 8:41 PM
I think many version managers make things unnecessarily difficult, especially if one hops from one repo to another. vmatch automatically uses and installs the right versions.
by jpc0 on 5/26/25, 12:39 PM
An app framework targeting modern C++ (>C++20) based on Boost.Fiber and stdexec( sender/receivers when C++26 releases)
I got inspired by a recent Java talk[1] from a Netflix employee about how they are building. I have the pretty strong opinion that async/await are implementation details and shouldn’t be exposed in APIs. They effectively have the compiler/runtime generate a state machine for you and you wouldn’t put a state machine in an api. Using C++ though we do have low level access to the CPU and there is no reason we cannot just use stackfull coroutines. This allows you to write literally synchronous code but it will defer blocking operations to an even loop and allow concurrency.
For GUI at least in the beginning Im thinking that I just spawn a native window and it’s up to you how you draw to that. I’m targeting something basic on linux myself which means it will probably just be direct calls to the wayland compositor to create a window. Getting a context etc will be up to the user since my idea would be in windows we use C# and on macos we use swift for UI. Documenting and building tooling around getting those languages integrated feels like it would provide more benefits than actually trying to reinvent the wheel in C++.
Likewise HTTP client will just wrap libcurl, for server side Im actually thinking of just calling into golang. HTTP servers aren’t the hard part, it’s all the infrastructure around it that is the hard part and well golang handles that better…
by mr-karan on 5/26/25, 10:29 AM
It's an open-source (AGPLv3), purpose-built log analytics UI specifically for ClickHouse. I've been using ClickHouse for logs and love its power, but found the existing UIs either too heavy, too focused on ingestion (which we already have covered with tools like Vector/Promtail/Fluentbit/Logstash etc), or not ClickHouse-native enough.
Logchef aims to be a lightweight, powerful log explorer that sits on top of your existing ClickHouse setup.
Key things:
- Schema-agnostic: Works with your existing ClickHouse log tables. No need to change how you store things.
- Dual Query Modes: Simple search syntax (e.g., level=error service=api) for quick looks, and full SQL for complex analysis when you need to dig deep.
- Lightweight & Focused: Single binary, easy to deploy. It doesn't do ingestion, letting you use best-of-breed tools for that.
- Team-centric: Built with multi-tenancy for teams and access controls for different log sources from the get-go.
I've got a demo running at https://demo.logchef.app if you'd like to try it out. It's still evolving, so feedback (especially from other ClickHouse users) would be super valuable!
Wrote an announcement post with a bit more detail: https://mrkaran.dev/posts/announcing-logchef/
by dd_xplore on 5/29/25, 10:29 AM
This is mostly a hobby/community help project. Majority of schools in my area are either not required with computers or they have lots of older ones with no funding to maintain our modernize them.
I'm trying to setup everything within VPN (either wireguard or tailscale). I'll most probably go with tailscale so I can use the subnet router.
The zero would would run the older RPiOS (32 bit) and would autostart a rdp session using xrdp client. It's pretty usable for mostly static work that students need in schools.
Each RPi unit would cost around ₹2500 ($29) which includes all the adaptors, power supply, sdcard etc.
On the server side I'm running two Ubuntu server OS VMs (inside proxmox) configured with XFCE4 and KDE.
The work is still in progress, I'm yet to iron out major bugs.
by prvnsmpth on 5/28/25, 2:40 PM
A platform for consulting aspirants to practice business case interviews.
Finding case prep partners is a major pain point for B-school students/consulting aspirants. Fortunately, frontier AI models are now good enough to function as surprisingly competent case interviewers.
by dwoobat2467 on 5/26/25, 6:09 PM
I'm going to add track guides to it, in the hope that, eventually, I don't have to google for track guides or anything, I can just have them all in one place for me.
I also want to make my own overlay by analyzing UDP data that the F1 games export, because I play on xbox, and my shitty laptop probably doesnt have space for simhub - but mainly because its cool. If i can figure out how, I might also integrate some AI thing to comment on improvements you can make to your driving based on the values too!
I'm a student still so the code organization at the moment is literal trash but im working on it - when it's at a point where even I don't cringe looking at it I'll post it here!
by burgerquizz on 5/25/25, 9:47 PM
here the games result so far: https://playcraft.fun
by kofta on 5/26/25, 9:10 AM
by sothatsit on 5/26/25, 1:04 AM
Funnily enough, players are much more excited about the achievements I am planning to add than they ever were about us solving this 4000-year-old board game. I guess you've got to pick your audience.
by oliwary on 5/25/25, 10:57 PM
We are up to almost 200 puzzles, with around 700 players per day. I've become much better at finding videos that work well as puzzles and am working on adding small quality of life updates.
by samsk on 5/26/25, 5:41 AM
by encom on 5/26/25, 2:23 PM
by reifcode on 5/26/25, 7:24 AM
My brother and his wife would like to have the same setup, but it’s definitely not user-friendly, so I’m trying to figure out a one-click way to set everything up so you can have your own private ebook catalogue on the go on the device of your choosing.
by allenleein on 5/26/25, 7:07 AM
A macOS app that uses your AirPods' sensors to catch bad posture in real time.
How it works:
Real-time tilt tracking – Your AirPods already have the tech Customizable alerts – Adjust sensitivity so it nudges you only when needed Prevent strain before it starts – Stop neck pain and headaches at the source
Demo:
by dhavaln on 5/26/25, 1:30 PM
StackAdvisor (A tech-stack advisor for MVPs and beyond)
- Understand the idea, team, budget, skillset, preferences (cloud services, language, frameworks), time to market, etc
- Generate smart questions and iterate based on answers
- Suggest rationale optimal architecture to begin with and implementation detail
- Include In Scope / Out Scope tasks, Week-1 to Week-12 plan
- Suggest low-code/no-code/AI-assisted tools for implementation
by oofoe on 5/26/25, 12:49 PM
Here's a presentation, includes examples: https://hg.sr.ht/~oofoe/candheat/raw/dox/intro.pdf?rev=tip
It's fairly free form, so you can do almost anything you want, but still structured and "legible" for computers and the person debugging it. I wrote it in a Lisp that compiles to JavaScript, but there's no reason you couldn't implement it in any language that supports associative arrays -- idea is that the language has most of what we want, just need to organize it a little. I think the idea is useful enough, that it's potentially useful for more than "just" games.
Repo with several more-or-less worked examples: https://hg.sr.ht/~oofoe/candheat
Recently used it for a Lisp Game Jam entry, which someone described as "Helltaker meets the Magic School Bus": https://oofoe.itch.io/class6
About 800 SLOC, compiles down to just under a hundred lines of JavaScript. This includes the utilities, the "engine" (keyboard handling, overlays, text and art composition, animation, sound effects, etc.) and the game logic itself (which includes a level editor).
by haiku2077 on 5/26/25, 2:13 AM
The DevEx isn't quite as easy, but I'm very happy with the final result. I don't have to run the overhead of a Docker Engine anymore, the network policy is far simpler to write and audit, and I can use the same logging tools for both containerized and non-containerized services.
The main hurdle has been file permissions for the mounted volumes for my data; Podman is rootless by default, which means you _should_ build your containers to run as non-root UIDs/GIDs and map them onto host UIDs and GIDs, then grant permissions for those host users and groups to access your data volumes. In practice, the easiest path is usually to run Podman in rootful mode, which is not a best security practice but avoids the difficult-to-troubleshoot file permission errors if you don't do the UID/GID mapping correctly.
However, unless you are really trying to optimize for overhead/performance, you should probably use k3s.
by kacesensitive on 5/25/25, 8:20 PM
by _kush on 5/26/25, 3:36 AM
I'm working on the next major update for LookAway. I'm improving the wellness reminders to be smarter - instead of nagging you on a timer, blink and posture alerts now only trigger when you actually need them.
Also adding productivity stats so you can see how taking proper breaks during prolonged screen time affects your work output.
by jhunter1016 on 5/26/25, 12:39 PM
My co-worker and I started this as a side project because we were building so many sample apps, proofs of concept, and engineering as marketing apps and we were feeling more and more locked into Vercel every day. So we wanted to go back to client and server separation and when we did that, we realized most hosting providers have moved toward server-side rendering as the default.
Client-side rendering isn’t for everyone, but plenty of people still make simple websites or even complex CSR web apps and don’t want to deal with fully integrated solutions like Vercel (and Netlify now to a lesser extent). Plus they want the flexibility of choosing different tools for their frontend and backend.
This remains a fully bootstrapped side project for us, but it’s gotten the most traction of any side project either of us has worked on.
by seanwilson on 5/25/25, 10:20 PM
https://www.inclusivecolors.com/
No AI or autogeneration stuff, more like an advanced editor that lets you tweak large sets of colors to your liking and test they pass contrast checks in advance before you start using them in your UI/designs.
by henadzit on 5/26/25, 10:03 AM
by martin-adams on 5/25/25, 11:06 PM
by mgl on 5/25/25, 8:12 PM
by nnq on 5/26/25, 6:04 AM
TINY attempt to fight the growing divide between "AI haves" and "AI have-not" with
- PREPAID access (+ bring-your-own-API-keys alternatives for more technical users) to premium models (ChatGPR plus and above)
- multi-channel access (SMS/MMS, Whatsapp + many many more) to give ANYONE, ANYWHERE, regardless of how non-technical they are access to absolute TOP/SOTA model advisors with easy access in their language + fact checkers + more (think african farmers looking for intelligence to grow best crops, former delinquents lookeing for rehavilitation, absolutely anyone deservers ACCESS TO INTELLIGENCE)
- zero lock-in - paid (freemium, but no adds) product, but 100% open data formats, sharing, import/export etc.
STAGE: prototype for personal use as a whatsapp bot+ webui, somewhat usable for us but still wip, next stage is building accounts and payments for EU access, next for US/international
by terrib1e on 5/26/25, 3:58 PM
I'm also available for other work/projects. I'm a full stack developer with a decade of experience working in government tech and real estate. Feel free to reach out!
by dainiusse on 5/26/25, 11:58 AM
It is a personal itch for my home Sauna. Lots of stuff to manage, but trying to work on consistency as a priority as I dropped the ball in many previous attempts.
by mazzystar on 5/26/25, 6:56 AM
If you'd like to play, visit: https://cike.app
by andrewrn on 5/26/25, 1:50 AM
Here’s a very early version for circuits: https://www.circuit-tutor.xyz
by fregonics on 5/30/25, 10:42 AM
This is currently just a prototype I'm developing, but I'm honestly eager to receive some feedback, as I'm not completely sure about the direction.
by keyserj on 5/27/25, 1:33 PM
The core idea is to ground discussion in a causal diagram that ties together problems & solutions, then each node/edge can have structured details (importance score, arguments, unknowns, relevant facts, etc.) to help clarify & refine the information. It also has some features for working with this information, e.g. comparing perspectives, using a table to evaluate tradeoffs between solutions.
Right now, you basically need to be a power user to get benefit from it, but I've got a lot of ideas for making it smoother, and I'm slowly working through them.
by quintes on 5/25/25, 9:56 PM
* prfrmHQ SaaS The modern way to manage performance reviews, set clear objectives, and ensure alignment across teams or individually — all in one place
see https://news.ycombinator.com/item?id=43538744 [Show HN: My SaaS for performance reviews setting goals and driving success] https://youtu.be/ygvKdgiKRj4?si=Q9ael-oCLEGKMIgN - Shows I can use AI and I've integrated into AWS Bedrock
- Shows I can integrate with Stripe for payments
* Consulting (Architecture, Strategy, Technology leadership and advisory) - I'm working on getting my consultancy started. If anyone wants the kind of skills I offer here let’s talk https://architectfwd.com
* Next SaaS - Starting a SaaS for managing core strategy and technology concepts.
by alprado50 on 5/25/25, 11:24 PM
by pfista on 5/26/25, 1:22 AM
Most ai apps are focused on reading and synthesizing data, but none are great at “write” actions across apps.
We’re focused on making the simplest and easiest way to use ai to control your apps and get things done
by nikodunk on 5/26/25, 3:25 AM
https://apps.apple.com/us/app/daily-optimist-think-positive/...
by trevinhofmann on 5/25/25, 11:19 PM
It was in the works for about a year, and I'm now trying to find ways to make it more marketable and useful.
The three main things I'm working on are:
1. GitLab support (GitHub only at the moment) 2. A demo on the landing page that doesn't require any sign-up. 3. Better "vibe coding" experience through the Chat interface for those who want it.
I built it with TypeScript on the front- and back-end, React, Node.js, and PostgreSQL. And I over-engineeres it, as one does, with a Redis cache and websockets to push the latest data to web clients so the latest info is always shown without needing to refresh. I'm using the OpenAI API right now, but I want to switch to local models when I can invest in the hardware for it.
Edit: https://mysticode.ai - Would absolutely love feedback.
by jprokay13 on 5/26/25, 12:45 AM
Slowtube was a huge inspiration for me (I’ve spent many hours with it), but I wanted a way to save my loops and continue my practice.
Saving is solely to the browser’s database right now.
Feedback is much appreciated!
by amorroxic on 5/26/25, 4:56 AM
- static (mp4) or live (hdmi) input, hdmi output
- a collection of effects (various distortions, color glitches, demoscene-like) applied over incoming video stream, maximum of 4 effects stackable on top of one another
- midi controllers support (controlling actions/params of effects via CC)
- modulation of effects params via LFO and audio events (bpm, kick, tonal detection)
- loading of .glsl shaders (eq. shadertoy.com)
- dynamic input resolution, output either 360p (rpi 4/5) or upscaled to 720p / 1080p (networks like SRGAN over Hailo / RPI or RK3588 with a Radxa 5B SBC).
Given a 2nd screen (timeline editor) would love to evolve this to something like a hardware editor, somewhat in the line of DAWs in the audio world. Most things are working with biggest challenge now being building a control surface (buttons, rotaries + associated oleds, etc) and attempting laying it all on a PCB, a process I don't know much about. If there's interest welcome comments and could elaborate more.
by duttish on 5/26/25, 8:07 AM
Been chasing pilot customer but it's few companies that only have android, so I'm picking up a mac mini after work to start working on the ios version, see what's possible with the various restrictions.
https://www.tricktrapper.com/ if anyone is curious, it's sparse of the details though. I'm working on a doc with all the technical info. The plan is to release it for free to private people, charge companies.
by casualmike on 5/29/25, 4:06 PM
You can drag and drop links from YouTube, Twitch, TikTok, or Kick.
You can add or remove streams, save mixes for later, and share them as links.
You can watch multiple videos at once in a wall-of-video grid and navigate quickly from one stream to another. It works best on a really big screen --phones aren't really supported.
So far, people seem to like it. I've made a lot of updates based on people's feedback since I announced it a few weeks ago.
by chimprich on 5/26/25, 8:38 AM
by tiondo on 5/25/25, 9:21 PM
Most safari booking sites are either outdated, opaque on pricing, or offer one-size-fits-all tours. We let travelers customize everything — dates, interests (e.g. big cats, birding, photography), travel style, and budget — and generate a full itinerary with lodge picks, activity suggestions, and accurate cost estimates (including seasonal pricing and transportation).
We also partnered with local operators so users can actually book what they see — not just get ideas. The goal is to make safaris more accessible and planning less overwhelming.
Still early, but if you're curious or planning a trip to Africa, I'd love feedback: https://greatriftsafari.com
by user070223 on 5/26/25, 9:02 AM
Sensing effectively by optimal (re)positining of the sensors (both number of sensor and physical placement) Think spatial sensor/data fusion, perhaps using more cheap sensors rather than few expensive
Building a model of the dynamics (based on both reasonable assumption about the system based on grounded physics, and dynamical interpretation of the sensor's data (PySINDy, DMD)
And eventually controlling the dynamics
Love to get your feedback, ideas, if you know about such products/business etc, how could it help any business, real life application, if it interests you, anything :)
by AaronAPU on 5/25/25, 8:29 PM
Fits into my “loudness series” suite of tools.
Have 3 more in development and then it’ll be on to the next series.
by ramoz on 5/25/25, 9:15 PM
While it may sound counterintuitive, the agents of today aren’t truly autonomous in that you need to really guide them and plan their actions well.
I believe this is true today, and will be even more true when agents are guiding agents.
We need new infrastructure for dynamic context management.
The answer is not as simple as “hook up your agent to an MCP that pull docs from the web” … also MCP needs its own revolution. I tend to use no MCP and prefer raw agent performance.
I’m evolving the simple concepts I built in my VS Code extension to address this. Nothing public now, but I and a few others use this everyday to feed parts of large codebases into Gemini (to build plans for Claude code, other coding agents): https://github.com/backnotprop/prompt-tower
by ostojan on 5/26/25, 6:05 PM
I treat the project as a learning opportunity. I’ve always wanted to work on some mobile apps, so Spark is my first step towards that.
by iamwil on 5/25/25, 7:39 PM
Reactivity can update the state of the notebook automatically, so you don't have to keep track of which cells to execute again. Side effects are managed to make it easier to reason about while maintaining reactivity and ability to interact with the outside world.
by schappim on 5/26/25, 2:52 AM
Most AI agents today are like smartphones without any apps. Ninja.ai makes it easy for anyone to install and use AI tools (MCP initially) that automate useful tasks, without needing to write code.
Think:
- One-click install for AI-powered tools (MCPs, Model Context Protocol servers)
- Automatically share relevant context so tools know what you’re working on
- “Recipes” that bundle multiple tools into workflows (e.g., check inbox → pull receipts → send summary to accountant)
Still early. Just launched our prototype. Bootstrapped so far, now fundraising.
If you’re excited about agents, app ecosystems, or making AI useful to normal people, I’d love to hear your thoughts.
by cmenge on 5/26/25, 6:11 AM
TenderStrike (https://www.tenderstrike.com) - AI for tenders in the construction industry. Fascinating how large and complex a tender can be, and how many there are. LLMs can be super helpful here, but there are many challenging parts to it.
Optivise (https://www.optivise.io) - product data optimization and translation using AI. We're focusing on a niche here, helping customers to onboard onto Mirakl-powered marketplaces, but hoping to expand to other marketplaces soon. I have spent the last years before that deep down the trenches of product data management and the traditional approach isn't fun :)
by vladris on 5/26/25, 12:29 AM
by stared on 5/25/25, 10:34 PM
Now it is early alpha, but you can already give it a try.
by csnate on 5/25/25, 9:06 PM
A binary static analysis tool that identifies vulnerabilities.
Right now, still just focused on buffer overflows. It can find some known CVEs and I’ve made several reliability improvements over the past month or so.
I think I’m going to expand to additional vulnerability types soon.
by scroogey on 5/26/25, 7:52 AM
I think I've finally gotten it to a point where it's usable and it seems to work on my phone as well. Hopefully baking in the lighting for the 3d models will give a small performance boost and I don't think it will be noticeable at all with the color changes and small movements.
I've also been going through the fastai course so hopefully I'll be able to put that to use on this project by generating midi files from audio! I might be underestimating the difficulties there but it should be fun.
by jblakely on 5/28/25, 2:26 PM
Is anyone still interested in cross platform development for desktop apps (Windows & macOS)? Yes, it's similar to Electron, but mine uses native UI components and there are no prerequisites to get started.
I don't want to violate any rules about self promotion, so if anyone is interested in checking it out, let me know and I can give you the link.
by wwall3r on 5/26/25, 4:22 AM
While the tourney isn't going year-round, it has a demo mode where you can play through prior years (with friends! Or with yourself on another browser/device). The chat feature is admittedly half-baked and is mostly there to get my wife to quit asking about it. Will try to get that fully done for next year.
I'm also working on another side project in Gleam just to check it out. That seems to be taking shape and once I get something decent I'll post it.
All of these are self-hosted via Coolify on a VPS, which has almost been a project to itself (I've had it with cloud providers). I've learned a lot and it was cool to see some other self-hosting solutions in the thread here.
by amterp on 5/26/25, 9:33 PM
Keen to hear people's thoughts and feedback! GitHub here: https://github.com/amterp/rad
The docs site also contains a guide for getting started, including some of the unique features: https://amterp.github.io/rad/guide/getting-started/
by puapuapuq on 5/26/25, 8:24 AM
I decided to develop a system that take requirements and some sample inputs as an input, then automatically evalute the outputs and fine tune the prompt.
by dimaom on 5/29/25, 6:33 PM
One thing I’ve found super helpful is showing users what’s missing — like skills or tools mentioned in the job ad but not in their CV. That little “gap check” makes the letters more focused and relevant.
Still building and testing, so if anyone here has thoughts on what actually helps in real job applications, I’d love to hear.
by spacechild1 on 5/26/25, 11:20 PM
AOO - a C/C++ library for low-latency peer-to-peer audio streaming and messaging:
https://www.soundingfuture.com/en/article/aoo-low-latency-pe...
"game over" - a musical 2D game engine: https://vimeo.com/389378372
by mikewarot on 5/28/25, 5:38 AM
Long ago, I came across Metamine here on HN, and thought it was magic... but there's a LOT of stuff in that code I don't understand.... maybe I'll do a Turbo Pascal version later? I do have code to do cooperative tasking under MS-DOS from the 1980s I wrote. ;-)
by Nhatnguyen02 on 6/1/25, 1:11 AM
by CommenterPerson on 5/26/25, 3:02 PM
I would like a phone app that would disable anything that a user doesn't actually need (not something like the existing google android feature that identifies apps that haven't been used in x months). It may ask a user what they actually need (for example, calling, SMS, a browser, etc.). The app would identify all the software/apps on the phone that are not required for the above needs and identify / disable them (say for example Location services). When you get a new phone one could start off with an absolute minimum.
It would need ongoing updates, as the app universe and surveillance technologies continuously evolve.
by kegs_ on 5/25/25, 10:26 PM
I think it works great! The problem is, I think it works great. The issue is that it is doubly-lossy in that llms aren't perfect and translating from one language to another isn't perfect either. So the struggle here is in trusting the llm (because it's the only tool good enough for the job other than humans) while trying to look for solid ground so that users feel like they are moving forward and not astray.
by adhamsalama on 5/25/25, 9:34 PM
I tried self-hosting Sentry recently and found out there are a lot of moving parts, which makes sense for their scale and use case.
I was wondering if I could build something small and not multi-tenant. So I started experimenting with writing a server (in Go) that collects OpenTelemetry data and inserts into Clickhouse, an API for retrieving data/statistics (P95 in a time range, etc...), and a frontend (React.js) that displays them. All of this in a single executable file (yes, including the frontend, but not including Clickhouse).
This is all very new to me so I'm learning Go, Clickhouse and OpenTelemetry at the same time.
by pizlonator on 5/25/25, 11:52 PM
Unlike most programs, which just work in Fil-C with zero or no changes, libffi needs to basically be rewritten. Instead of using assembly for reflectively crafting calls it needs to use the Fil-C zcall api. And instead of JITing closures it needs to use the Fil-C zclosure_new api.
Should be fun
by loufe on 5/26/25, 1:59 AM
by rjmunro on 5/25/25, 10:27 PM
by cddotdotslash on 5/26/25, 7:53 PM
I got tired of using the AWS console (the UI is inconsistent across services, resource-heavy, and loaded with things I don't need)
I've built a handful of features, mostly to scratch my own itch, including:
• Multi-region mode (select 2+ regions and see all your resources on the same page)
• Automated diagrams (tree-style resource relationships)
• Easy export to CSV
• Reference matrix of AWS service/region availability
The best way to explain it is with a demo, so I built a demo version: https://app.wut.dev/?service=acm&type=certificates&demo=true
by kanishkalinux on 5/26/25, 9:15 AM
Link: https://github.com/kanishka-linux/inquisitive
I've been working on this on and off since last couple of months to consolidate all of my digital knowledge-base like notes/links/pdf etc with the help of LLM/RAG. It is fully self-hosted with very minimal setup instructions - which should be easily installable by anyone on their local machine.
UI is very bare minimal. It still has some rough edges when it comes to the UI part, but usable. I've been regularly using it since sometime, and it works well (atleast for my use case) when it comes to searching and organizing personal knowledge-base.
Feel free to try it!
by cornfieldlabs on 5/26/25, 12:35 AM
A place to dump all my thoughts in text to people I already know.
No photos, no clout-chasing, ego boosts, infinite scrolling, findable profiles etc.
I created a waitlist page (code only) using Gemini 2.5 and hosted it in cloudflare pages using D1 for database.
Waitlist page is here: https://waitlist-tx.pages.dev
Here's the full feature list for the initial version I am building where you can leave comments anonymously:
https://docs.google.com/document/d/1wkxk01C8ePEQ2jkwwt6p5Z7V...
Any feedback is appreciated.
by nickandbro on 5/26/25, 8:18 PM
It's a site where I hope to make learning vim more fun by competing with a AI bot (powered by reasoning models) when solving challenges. Looking to hopefully complete the project very soon.
by andjar on 5/26/25, 1:18 AM
I'm building a simple outliner focused on core functionality: hierarchical notes, backlinks, file attachments, and todo management. The goal is portability and longevity - it runs in phpdesktop as simple php, js and sqlite without compilation or complex frameworks.
I have been a happy user of LogSeq, but the current major rewrite, intended to enable real-time multi-user collaboration, introduce a lot of complexity and I am now more worried about the longevity of that project.
Built with assistance from Cursor, Gemini, and Jules, so vibe coding. Still early stage but functional for basic outlining needs.
by dalemhurley on 5/25/25, 9:46 PM
by siddmax on 5/26/25, 7:49 AM
First time trying vibe coding as a staff SWE at F, so it’s been a lot of fun to try things out. 100K+ lines of code without a single line myself. Working on a new process to one shot the mobile app from scratch (even deployment). Learned a lot that I apply to my day job. Hoping to make an agent builder next!
by HarshitDoshi on 5/26/25, 5:01 AM
https://shunyaek.se is my agency, "shunyaek"'s home-page. shunyaek refers to "shunya", meaning zero, and "ek", meaning one, in Hindi, my native language. It refers to the binary building blocks of the software & digital world. The landing-page is still under construction. Always open to feedback and crticism.
by nichol4s on 5/26/25, 11:52 AM
For example, if during the conversation someone asks, "tell me how this API call works". It should be able to. 1. find the right API document (requires search) and then 2. 'retrieve' that API document. But if that API document also requires other content (eg Authentication is separate) that will require ANOTHER roundtrip. So I'm currently trying to figure out what the best flow would be for this.
by monsieurpng on 5/26/25, 3:00 AM
Feel free to download here: https://learnmathstoday.com/2025/03/02/learnmathstoday-app-i...
by ezhil on 5/26/25, 3:03 AM
by axelthegerman on 5/26/25, 3:50 PM
[freshbuzzer.com](freshbuzzer.com) does that and much more!
by sagering on 5/25/25, 7:45 PM
Feedback, suggestions or contributions are very welcome! :)
by hluska on 5/26/25, 1:59 PM
Which just goes to prove, if you’re slightly demoralized it’s a side project. :)
by fratimo66 on 5/26/25, 2:13 PM
Turns out people really like it, especially for food photography :)
by benstigsen on 5/25/25, 9:58 PM
I am hoping this will be the way in which I write most of my future scripts and projects.
[1]: https://luau.org/
by patcon on 5/26/25, 4:48 AM
https://patcon.github.io/polislike-opinion-map-painting/
Am also working on a python library to generalize the pipeline of Pol.is, so that more projects can use its features without recreating from scratch:
by mdolah on 5/26/25, 6:12 PM
https://github.com/MahmoudDolah/linux_hwinfo64 https://pypi.org/project/linux-hwinfo64/#description
I started out by copying features of HWINFO64 (which does not run on Linux). I'll probably keep hacking on this for a bit. It's been interesting to dig deeper into how hardware and software interface
by mdh on 5/30/25, 9:22 AM
by Brystephor on 5/25/25, 11:09 PM
by Tanzirul on 5/25/25, 11:32 PM
The idea is simple: instead of blasting you with every keyword mention like F5bot, Replyhub filters for posts where people show real buying intent. These are posts where they ask for recommendations, compare products, or look for solutions.
It also suggests context-aware replies and helps collect leads from people who show real interest.
If you want to reach niche communities where people are actively discussing products, it might be useful.
Would love to hear feedback or questions from folks here.
by 6stringmerc on 5/25/25, 10:05 PM
A totally bootstrapped, professional services undertaking with no investors needed. The value is in the knowledge acquired over a decade plus in sales support roles and learning about an underserved, viable market.
by ciccionamente on 5/26/25, 8:21 AM
by vertix on 5/26/25, 8:06 AM
Fully open hardware and software, including ML, so that you can start building useful robots immediately
by mikeytown2 on 5/25/25, 9:48 PM
CLI Meshtastic flasher that works well. No internet mesh networking sounds awesome; just the bandwidth is extremely limited
by cerlo_team on 5/26/25, 2:08 AM
Most tools we’ve tried are either too bloated, too expensive, or limit you to one or two accounts per platform. We’re aiming for something simpler: connect multiple accounts (free up to 3 per platform), schedule posts, organize media, and view basic performance stats — no AI or calendar features yet.
Still early, but the core workflow is live and usable. Would love to hear how others here handle multi-platform content publishing, especially if you’re doing it without a full-time team.
by harlanji on 5/27/25, 4:13 AM
Mostly not open source, contrary to how I have historically operated. I think I'll drop them open source as time goes on and money matters less. Kinda feels like I might make more money just open sourcing it all because of good karma tho.
It's all in Python because that's what I could get on the $200 Windows laptop I built it with, I kept it in S Mode as a challenge because I want it to run as constrained as my customer personas might experience.
Present stage is finding those personas and slicing the wares up into offers/products.
by primitivesuave on 5/25/25, 11:44 PM
While I love the official TypeScript handbook, it's not easy to play around with the code examples or approach it as a beginner. I started working on a complete TypeScript tutorial that also showcases some advanced use cases. All the code examples run in the browser, and there are some neat visualizations that clearly show what the type system has picked up.
I've been trying to fix some of the performance issues, finish writing all the content, and adding documentation before making the GitHub repository public - right now the page can hang when loading a long tutorial.
by Levitating on 5/25/25, 11:47 PM
I am also working on a web frontend for rrdtool (for graphing collectd statistics): https://github.com/LevitatingBusinessMan/collectrack
And a wayland bar that is configured via a Ruby DSL: https://github.com/LevitatingBusinessMan/rubybar
by VladVladikoff on 5/25/25, 11:22 PM
by _luiza_ on 5/26/25, 9:25 AM
In principle, using my current free time (& not only tbf) to figure out a way to fund a few positions on full-time study / audit of emergence, specifically emergence of consciousness in this case.
Setting a basis in semantics & wanting to optimize as sense gets un-sparsed.
Hmu if interested to figure out some of these together (agiornot@gmail.com)
by dennis16384 on 5/25/25, 9:20 PM
It's been 6 month since our first appearance on Show HN [1], and I'm working with first free users on bugs, improved workflows and UX, geocoding, solver features, future mobile app etc. etc.
We officially crossed the limits of 1500 stops per optimization with some waste collection guys, all still running fully client-side in the browser.
by pravj on 5/26/25, 7:12 AM
I stopped writing right after college when I moved to full-time work, but I want to change that now.
Just as I started, I realized the website was hosted on Forestry CMS, which has been discontinued, so I will have to figure out how to maintain the URL structure, etc. (this is important since there are a few popular/top-10 pages for certain queries/searches [1]).
[1]: https://hackpravj.com/blog/solving-semantris-opencv-word2vec... (for "Google Semantris")
by stonecharioteer on 5/26/25, 2:20 AM
I want to build a learning platform powered by deliberate practice and AI. To teach how to program from a practical perspective.
I am also trying to kick off a podcast for folks in tech at Bangalore. I hope to start it in June.
by Bramhoven on 5/27/25, 6:56 AM
The idea is to bring more structure and insight to self-improvement without making it feel like a chore.
I’d really appreciate any thoughts on the concept or the site itself: https://proflect.io — does it make sense? Anything unclear or missing?
by sanswork on 5/25/25, 8:53 PM
by Jean-Philipe on 5/26/25, 12:47 PM
because I got frustrated with meetup.com
by lukehollis on 5/25/25, 9:58 PM
I was working on world models / generative environments but without the training data available as an independent researcher, ended up focusing on building with existing geospatial data.
The same architecture of the '24 Genie paper's dynamics model is instead trained on historical data for risk analysis and creating a heatmap in the 2d map. I'll try to adapt this for a more generalizable urban mobility model as well.
by sandeepkd on 5/26/25, 5:28 PM
The authentication space becomes too complex above a certain company size, the concepts like admin user, service user, are not always supported as first class concepts. In a nut shell, just trying to solve the problems that I have personally come across the last decade and more in the industry in authentication space
by conditionnumber on 5/25/25, 8:52 PM
>90% of the work to replicate academic research in quant finance is data collection and cleaning. Academics have WRDS, Bloomberg, and research assistants. Others need to find cheap vendors of acceptable quality. Write scripts to fetch and clean. Actual algorithms and analysis (the fun part) are a tiny fraction of the total effort.
Project objective is "good enough" data collection & cleaning for daily frequency equities, futures, and FX. In that order. Annual data budget target < USD 500. Cheaper is better.
by cryptoz on 5/25/25, 8:53 PM
You can build webapps very quickly, especially AI-enabled ones, and deploy them on a subdomain. Other users can sign up and use your webapp, and any tokens they use will be billed to them and you will get a large cut (80%) of the margin earned on the tokens billed - as I bill 2x OpenAI API token costs to create this margin.
So ideally you can validate your idea by rapidly building a prototype and evening earning revenue to boot.
by austin-cheney on 5/25/25, 7:58 PM
by yboris on 5/26/25, 3:09 PM
https://github.com/whyboris/utilitarianism.net
Built this website with Hugo and it's been one of my favorite (and best) projects of my life. Has PDF generation of articles and a static low-bandwidth instant search of the whole website. Very happy with the build system too (using just)
by brainwipe on 5/26/25, 3:40 PM
- Building plastic self organising maps (Lang 2002) using Python CUDA build to parallelise the more expensive bits. Also fancy building the directed graph half in Unity 3D.
- Also doing some data engineering pre-training and AIIA at work but no deets, sadly.
by ChrisMarshallNY on 5/25/25, 11:08 PM
I have a couple of other apps that I have plans for, as well. If I get sick of playing traffic cop, with the phone app, I may take a break, and work on them.
by kazinator on 5/26/25, 6:20 PM
It is just now a matter of, cut it now, or include yet one more thing?
I started writing the RELNOTES two weekends ago, yet more stuff started going in.
In the semantic versioning scheme selected for this project, 300 means that it is literally the 300th public release.
There are over 140 new commits and 299 was all the way back in February.
It's going to be very exciting (for the small number of people that know about this project, now nearing its 16th anniversary).
by cookboox on 5/26/25, 1:07 AM
by weakfish on 5/25/25, 10:22 PM
A block-based TUI note/task application using the Charm tools. I know there’s a billion note apps out there, but none fit my mental model, so just hacking my own.
Goal is to have a system of dumping info in and letting organization naturally rise from tagging.
Each tag has its own page that aggregates all blocks tagged with it, and can have a custom page layout depending on the defined “type” of the tag I.e. a person, project, etc.
Tasks are also first class citizens and can be aggregated with dependencies on other tasks.
by nico on 5/25/25, 8:35 PM
Also having fun one-shoting or few-shoting, little games and interactives:
by yeutterg on 5/26/25, 12:58 AM
This builds on the back of our popular Bedtime Bulb light bulb - much-improved v2 coming soon: https://get.bedtimebulb.com
by bishopsmother on 5/25/25, 9:09 PM
[0] https://blog.walledgarden.ai/2025-05-20/wabbit-s2-welcome-to...
[1] https://www.sesame.com/research/crossing_the_uncanny_valley_...
by Bismayy073 on 5/29/25, 8:13 AM
The interesting challenge was balancing response quality vs speed while handling real-time audio processing from browser tabs.
Use case is interview help, but could work for other real-time conversation assistance.
Feedback welcome for interviewbee.ai
by thingsilearned on 5/26/25, 12:55 AM
https://github.com/davefowler/prosemirror-suggestion-mode
and now I'm building it into my reading and note taking app for a better interaction of AI giving assisted edits, most similar to the Patchwork project from ink and switch https://www.inkandswitch.com/patchwork/notebook/07/
by leansensei on 5/25/25, 10:00 PM
by hxii on 5/26/25, 4:21 PM
Getting the parts that I absolutely HAD to implement while not descending into feature-creep hell (driven by my own curiosity) was challenging.
Then again so were the self-debates about the syntax and documentation.
Maybe one day I’ll learn Go or something, but until then - Boku is my yaml-based task runner written in Python.
by DeonRob on 5/26/25, 7:26 PM
Sashi is an open-source workflow automation system that lets you turn real backend functions into reusable, shareable workflows—created with AI, validated by code, and executed from a secure dashboard.
*What makes Sashi different?*
- *Real Code, Not Just Integrations:* Register your actual backend functions (Node/TypeScript) with a lightweight library. Sashi uses Zod for type-safe validation and auto-generates UIs based on your return types.
- *AI-Generated Workflows:* Describe what you want in plain English—Sashi’s AI chat interface builds and validates multi-step workflows, connecting your registered functions in powerful sequences.
- *No Admin Panels to Build:* Sashi provides a secure, auto-generated dashboard for your team to run, manage, and visualize workflows. No more custom admin UIs or glue code.
- *Secure by Design:* All execution happens server-side, with scoped API tokens and session management. You control your stack and infra—no vendor lock-in.
- *For All Teams:* Support, product, DevOps, and engineering can automate internal tasks, trigger backend logic, and collaborate—without waiting on developer cycles.
*How it works:*
1. *Register Functions:* Use the Sashi library to expose backend functions with type-safe schemas.
2. *Create Workflows:* Use the chat or flow builder to compose workflows—AI helps connect steps and validate data flow.
3. *Execute & Visualize:* Run workflows from a central dashboard, see results in real time, and auto-generate UIs (tables, cards, graphs, etc.) based on your data.
*Example Use Cases:*
- Support: “Refund a user” or “Reset password” via dashboard, no SQL or custom tools.
- Product: “Send weekly usage report to Slack” or “Find users who abandoned signup.”
- DevOps: “Spin up preview envs” or “Rotate API keys” as repeatable, auditable workflows.
*Try it or learn more:* https://www.usesashi.com/
by NoTranslationL on 5/25/25, 8:30 PM
Reflect - an app to track anything and analyze your data, including a feature to run self-guided experiments [0]
Later - an app to schedule non-urgent tasks and ideas, with an SRS-like scheduler to punt items [1]
[0] https://apps.apple.com/us/app/reflect-track-anything/id64638...
[1] https://apps.apple.com/us/app/later-set-intentions/id6742691...
by singh273 on 5/28/25, 11:41 PM
by estsauver on 5/26/25, 8:03 AM
I'm also making a picture book generator for my kids. I want to be from some pictures, a description of a story, and will then pop out a ready-to-print picture book for them.
by sage76 on 5/25/25, 11:11 PM
by leslielurker on 5/25/25, 9:39 PM
by wonger_ on 5/26/25, 12:09 PM
by jkoff on 5/25/25, 9:58 PM
I'm building Infinite Pod, a web app that generates language learning podcasts tuned to your individual learning goals and level.
It's based on the principle of language acquisition through comprehensible input, as described here: https://www.youtube.com/watch?v=NiTsduRreug
It's still a bit rough, but feels magical in my own testing so I wanted to make it available to others.
by radsj on 5/26/25, 4:37 AM
Working on solving my own pain point, around saving web pages to apple notes. While there are workarounds like singlefile and apple shortcuts, it ends up being a multi-step process, and limited search functionality within apple notes.
Also, experimenting with using texting to manage family calendars: https://textconcierge.ai/
by rubyfan on 5/26/25, 3:22 AM
by vldszn on 5/26/25, 9:01 AM
It supports multiple languages, currencies, European VAT deductions, and more.
Originally made it for myself, so it’s a bit of a personal tool — but I hope others find it useful too :)
https://easyinvoicepdf.com/en/app https://github.com/VladSez/easy-invoice-pdf
by steida on 5/25/25, 11:21 PM
by fullstackchris on 5/26/25, 10:31 AM
Basically, event sourcing for the IDE - build entire software lessons and course videos without actually recording anything.
You can get up and running pretty quickly at our studio https://studio.codevideo.io
Or locally with the CLI tool: https://github.com/codevideo/codevideo-cli
by jsemrau on 5/25/25, 10:08 PM
Initially I thought there is a use-case in finance, but the barriers of entry are incredibly small and the value add is not that large.
Currently, there seems to be a lot of traction in code generation (Cursor, Lovable, et al), but I have not seen that work on a useable code base/workflow.
by kingo55 on 5/25/25, 10:43 PM
I've always enjoyed the farm-to-table concept, but I find it really hard to identify trustworthy companies. Wine has been done to death, but I feel extra virgin olive oil is currently underserved.
by cosbgn on 5/27/25, 8:18 AM
Rispose is the first tool I build which isn't tightly coupled on another provider.
I can easily swap OpenAI for another provider or even host my AI. This IMO is something really interesting and something before was relay hard to achieve (most tools are built on top of bigger platforms).
by huksley on 5/26/25, 8:12 AM
DollarDeploy is like Coolify and dokploy but with better UI, no docker containers and fast deploys for full stack apps, NextJS, react
by upmostly on 5/26/25, 1:09 PM
What's different with Hypership is that your apps get user auth, page view analytics, event tracking, logging, and an admin dashboard all set up and ready to go out of the box.
It's also a much nicer platform for devs because we host VS Code in the cloud for you as you're vibe coding. That means you can manually edit the code as you go.
It's basically what web development should be in 2025.
by dzink on 5/26/25, 3:54 AM
by quinto_quarto on 5/26/25, 6:30 PM
you can do all the usual stuff (save any link with one click, tag your stuff, add notes to each link). i'm not sure what to do next with it, open to suggestions.
i also made a IMDB of the creator economy, a free database of indies making cool things: https://indieworld.io/
by Tsarp on 5/26/25, 9:43 AM
Inspired by secure password managers like Bitwarden, goal is to reduce detectability, avoid CAPTCHAs, and mitigate common fingerprinting pitfalls.
The idea is simple: leverage the trust your browser already has.
by f3b5 on 5/26/25, 11:21 AM
by hummusisdata on 5/26/25, 8:17 AM
So I started my blog Cooking Data, and slowly, using Vibe Coding, I am building tools to translate the philosophical conversation into actionable tools. I am always happy to learn and discover how the team handles the data strategy and answers the business goals with it
by kanodiaashu on 5/26/25, 6:13 AM
I’ve not been satisfied with how AI is bolted on after the fact — reading in one place, notes in another, and AI as a separate assistant. So I’m combining them.
It works over EPUBs, PDFs, and HTML — you bring your own sources (BYOS), and it turns them into a AI scaffolded, structured, hierarchical reading experience.
Already using it for both reading and deep research style deep dives, and its already way better than my otherwise broken workflow!
by sircastor on 5/26/25, 2:32 AM
It’s been a very instructive process and it’s shown me where the strengths and weaknesses are.
As an aside on that. I don’t think AI is going to replace Developers. I do think it’s going to be a rough couple of years while businesses try everything they can to make that happen. And that’ll probably be disastrous for everything.
by tha00 on 5/25/25, 8:29 PM
by crabsand on 5/25/25, 10:40 PM
In the long run, writing a gui for https://github.com/iesahin/xvc and Git.
by chrisvalleybay on 5/26/25, 6:29 AM
I still don't have a landing page, but if you are curious you can test it out here: https://lindon.app
If you do test it out you can see all keyboard shortcuts with 'Shift+?' and you'll have to enable VIM-support in the settings (Cmd + K to open).
by r0x0r007 on 5/26/25, 10:03 AM
by scottishbee on 5/26/25, 9:17 PM
Every startup/company I work for I've had to jury rig ways to get new product ideas in front of users. Google sheets, Grafana boards, Chartio (RIP), etc all to go "SQL query" -> "user-accessible website". For some reason Retool doesn't have a "make externally available to these people".
So I'm playing around with how to build a generic interface between db and user access panel.
by yakshaving_jgt on 5/25/25, 11:10 PM
by greenie_beans on 5/27/25, 11:35 AM
by aniket_920 on 5/26/25, 4:03 AM
Here every users votes are weighted, means it's not necessary that every user votes count equal to one, may be it will be less than one or in negative also.
Algorithms is very complex and I'm figuring it out.
Currently at ideation stage and soon start the development.
I'm working on this idea to make sure my platform present a clear view about the topics from general public's.
by coro_1 on 5/25/25, 10:23 PM
by Yabood on 5/26/25, 1:03 PM
by woutr_be on 5/26/25, 4:15 AM
For a next project I'm exploring building something that let's you quickly embed charts. Similar to what Google Charts was, but I still haven't found a good alternative.
by Havoc on 5/25/25, 10:47 PM
Software wise doing proxmox + nixos LXC
by yoz-y on 5/26/25, 6:30 AM
It started as a bunch of spiral bound notebooks. I made an iOS prototype but realised it’s so niche that it would never get approved. Made a PWA that I’m currently using for about a year. Now I’m building a second version which I finally plan to share.
The reason I’m making it is that I realized that any other style (mainly, tabular input, dropdown boxes, input fields) just doesn’t fit my routine.
by getgalaxy on 5/26/25, 2:42 PM
We think theres a huge gap in tooling for devs associated with writing SQL. We've reimagined the SQL editor to make it modern, fast, beautiful, and have a AI copilot that actually works. Importantly, we've also added sharing and collaboration in a Postman / GDrive nature so you never have to share queries with your team via slack or notion again.
Check us out and ping me on LI. getgalaxy.io linkedin.com/in/garrettawolfe
by nikkwong on 5/26/25, 1:04 AM
by recsv-heredoc on 5/25/25, 10:04 PM
We’re solving the problem of “How can agentic AI interface with legacy and existing business systems.” - if you’ve got a boring job and are tired of filling out forms in business software or swapping between 10 different systems, convince management to let us come and have AI do it for you.
by ChicagoDave on 5/25/25, 9:51 PM
- mach9poker.com: incorporated startup developing a poker tournament training app for novices and unprofitable players. Looking for UX/product designer co-founder.
- policyimpact.org: A journalism site for highly vetted articles responding to actions of the current U.S. administration and other import political vectors.
- sharpee: a new interactive fiction platform built in Typescript
- bsky.poker: root domain for poker community to have nice handles on BlueSky
Happy if anyone wants to pitch into any of these projects.
by zoidb on 5/26/25, 9:03 AM
by Tsarp on 5/26/25, 9:51 AM
Spent a lot of time on low level hardware libs to roll out my own version of VAD, grammar correction and stitching segments.
Faster than the hosted dictations tools thought it runs locally and a lot more control in terms of custom vocabulary.
by bitbasher on 5/27/25, 10:23 PM
For example, you can research the best genres for indies by copies sold, numbers of released games/saturation, etc. You can also find youtubers that have streamed similar games/genres and get their contact info.
by dhuan_ on 5/25/25, 10:10 PM
the process of creating APIs for testing and automation should be as easy possible. the tools that exist nowadays aren't good enough, they require you to use their programming language of choice or complex procedures for a task that should be simple. I built mock to try to solve that and still continue to maintain it.
by ata_aman on 5/25/25, 9:33 PM
It's a file explorer where it embeds your local file structure so you can use natural language to search your file system.
Started off as a local inference/vector-db only project last year and now also using cloud inference/vector-dbs for faster processing.
You can also use "agent-mode" to organize your files/folders, create items, move, copy and save content to disk directly from chat.
by rjprins on 5/26/25, 6:58 AM
by Two_hands on 5/26/25, 7:39 AM
Built using python the app alerts you to when someone is looking at your screen, using locally running deep learning models + your webcam.
I developed the app when I felt uncomfortable with working in public spaces, I wanted EyesOff to give an extra barrier of security to shoulder surfing scenarios
by tmilard on 5/25/25, 10:25 PM
This has been my WE project for a long time. But it's only working really now.
by csomar on 5/26/25, 2:33 AM
Building Tools for working with Git and managing your code base. Initially started with an online merge conflict tool integrated with GitHub. Now working on a Code Owners CLI https://github.com/CodeInputCorp/cli to improve on the current Code Owners functionalities on GitHub.
by drchiu on 5/26/25, 3:24 AM
It’s an email marketing tool, self-hosted, for use with existing ESPs like Postmark, Mailgun, Amazon SES, etc.
Been working on this since September 2024, and have been consistently hacking away at it weekly: https://sendbroadcast.net/changelog.
by egonschiele on 5/25/25, 10:48 PM
by rgyams on 5/25/25, 8:32 PM
by fathermarz on 5/25/25, 11:53 PM
Would love some feedback overall and suggestions: https://phended.com
by madduci on 5/26/25, 5:31 AM
Recently we have released everything as open source, even the code for setting a local development Kubernetes cluster using OpenTofu
by antoferra on 5/26/25, 3:21 PM
Download link: https://apps.apple.com/it/app/timecap-limit-screen-time/id67...
by elananandhan on 5/29/25, 5:02 AM
by yuvadam on 5/26/25, 8:23 AM
by kemyd on 5/25/25, 9:36 PM
For the last few weeks, we have been working on catching up on features for vibe coders (prompt -> project), but now we are back to our strengths (visual editor and new beautiful UI libraries for Tailwind CSS, Bootstrap, and more).
We realized there are just too many apps for vibe coders, and it would be better to work on something unique that we are really good at!
by justacoder12 on 5/26/25, 5:09 PM
by simlevesque on 5/25/25, 11:40 PM
The idea is that instead of running Nominatim which is costly you can just query Parquet files over the network.
Instead of a cluster of PostgreSQL servers all I need is a bunch of static hosting holding the dataset that's around 1Tb.
Send me an email if this interest you, it's in my profile.
by adityapurwa on 5/25/25, 10:57 PM
An app that can turn anything into adorable stickers. In my region, people uses WhatsApp a lot, and there's this ability to create custom stickers. So we uses a lot of stickers on a conversation.
by bonniesimon on 5/27/25, 1:13 PM
Existing alternatives: Clay, Dex.
by bwb on 5/26/25, 1:53 PM
Excited about it, as I've been wanting something like this for a while :)
by Hyperized on 5/26/25, 5:36 AM
https://latencychecker.com does just that, because opening a ping prompt on a mobile device isn’t a thing without installing apps.
And it gave us a good excuse to stretch the legs and dabble in a little frontend development.
by userundefined on 5/26/25, 2:42 AM
by andoando on 5/25/25, 8:59 PM
On one hand the idea seems so simple and intuitive (Define patterns (like 3 red blocks to the right), combine patterns ( 5 up * 3 red right), use patterns inside patterns (each block is a square), but implentation wise I keep running into so many intracies and I want it to be perfect so it's been kind of tough and slow.
by hbroadbent on 5/26/25, 7:55 AM
It's starting to pick up! Which is cool — a few new users are trickling in each week. I think it's about time I invest more time in marketing now though (which is probably what I should be doing now instead of posting on HN!)
by emadm on 5/25/25, 10:48 PM
Stuff that should be open source, open data
Made state of the art datasets, health models, research systems & agents so far @ www.ii.inc but the plan is ai first open source full stack systems for every regulated sector
Have a distributed ledger announcing soon to tie it all together and create a flywheel so more folk can get access to ai
by rasulkireev on 5/26/25, 3:26 PM
Thinking about future monetization efforts and found myself liking Posthog approach for that.
Host for large companies looking to automate marketing, yet leaving it free (self hostable) for smaller companies that can't pay.
by 35mm on 5/25/25, 8:19 PM
by darkhorse13 on 5/26/25, 12:03 PM
by ubavic on 5/25/25, 10:03 PM
by akkartik on 5/25/25, 8:14 PM
by cckolon on 5/25/25, 11:39 PM
https://bearingsonly.net - a submarine combat game in the browser.
by agcat on 5/26/25, 6:22 PM
by 1270018080 on 5/25/25, 8:54 PM
by guywithahat on 5/26/25, 5:57 PM
by turbotim on 5/25/25, 10:51 PM
by simmschi on 5/27/25, 12:54 PM
by cloud8421 on 5/26/25, 9:44 AM
It integrates with MusicBrainz and Last.fm, so that I can easily add records, scrobble them, and search/track stats.
I want to figure out how to track availability for records I want to buy, which would involve interacting with marketplaces.
Using Elixir, SQLite and Phoenix LiveView, which is what I know best and lets me focus on UX and UI.
by c0nstantine on 5/26/25, 6:21 PM
by thom on 5/25/25, 9:48 PM
by felipesabino on 5/26/25, 8:31 AM
The result was very satisfying, https://rss.sabino.me - a rss new aggregator, with a summary to save some time on news that I do not want to read, but am mildly interested.
I will not say the same about the code quality tough
by AaronKosovich on 5/26/25, 7:11 AM
by pugworthy on 5/26/25, 2:28 AM
And get my MG Midget back on the road.
by muconto107 on 5/27/25, 4:57 AM
Queue management AI agent where clients interact using Whatsapp/sms for appointment and reservation;
School management system that integrates AI for reporting and teacher/student/staff task s assistance;
A cloud based remote execution platform to run python(and other languages) scripts in a isolated virtual machine.
by immibis on 5/26/25, 9:15 AM
Nothing useful, like always. A non-computer project for the next week or so.
> Any new ideas that you're thinking about?
A few, but I'll never do them. It would be cool to offer free hosting services to projects (in lieu of places like Github and Fandom, and in the spirit of Miraheze) but in the current legal environment, feels like I'd be risking my life.
by AttentionBlock on 5/25/25, 10:44 PM
by m5x5 on 5/26/25, 6:50 AM
by DylanSp on 5/27/25, 9:14 AM
by fieg on 5/26/25, 7:25 AM
This is a lightweight alternative to my earlier startup attempt Strateamic [2].
[1]: https://talkcue.app
by pul on 5/26/25, 8:55 AM
Beta is open at https://beta.nslookup.io (registration wall for now, but I'll remove it at launch).
by bob1029 on 5/25/25, 10:54 PM
There's no reason I should have my browser tabs crash when I view a pull request involving more than 100 files. The page should already have been generated on the server before I requested it. The information is available. All that remains are excuses and wasted CPU cycles.
by kaiherng on 5/25/25, 10:47 PM
by wcedmisten on 5/25/25, 9:15 PM
I've also used the data corrections submitted by users to contribute over 3,000 edits back into OSM!
by Jotalea on 5/26/25, 3:12 PM
by chrischen on 5/26/25, 7:02 AM
I implemented a simple first-come-first-serve event RSVP system, and from the event RSVPS it will pull the players into an event management UI from which the organizer can place players into the available courts and arrange matches balanced by level automatically.
The app does ratings calculations fully locally in the browser so it not only works offline but every new match is based on the last rounds' updated player ratings immediately. There is an option to submit matches to be persisted in my hosted service (pkuru.com) so player ratings are retained. The goal isn't to rank players like existing popular systems (DUPR, UTR, etc) but to do matchmaking within an organized session, or before a session.
The web UI and session management app, which is tentatively named FairPlay, is open source on my github. The version I host at Pkuru.com is free to use as well, and currently used by the #1 club (by average skill) in central Tokyo, as well as the highest level club in Thailand.
I don't plan to commercialize it as it is merely a side project but I can't promise the service will always be free to use due to potential operating costs at scale. However, the front-end UI including the session manager is and will be open source.
There's no fundamental reason the system cannot be easily adapted to other sports such as volleyball, soccer/football, or even a 100-person race since the rating system was designed specifically for team based competitions. It can do everything from ranking players in game like PUBG (100 teams of 1-4 players), to 5v4 futsal (asymmetric teams), to classic 1v1 duels as well.
If any clubs or even businesses are interested in using it to manage open play or rec play sessions where people can just drop in and out and be put into a balanced match of the appropriate skill level, please feel free to contact me. The hosted version on pkuru.com is free to use, but I can also assist in my free time on a volunteer basis if you wish to adapt the front-end open source code.
by motohagiography on 5/26/25, 7:35 PM
- applying ML/AI to some security research with some defence or product possibilities
- impementing a test for a quant'ish trading idea that just won't leave me alone
- preludio en mi menor by augustin barrios mangore
- piaffe to canter transitions on a horse
- still figuring out what I actually want to do, but mostly who I want to make something for
by tsbischof on 5/27/25, 11:00 AM
by Andi on 5/27/25, 1:03 PM
by ryukoposting on 5/25/25, 11:28 PM
I'm going to plug a couple phones into it, but the main goal is to get all my old computers to talk to each other using their modems.
by p-s-v on 5/25/25, 10:44 PM
by anfractuosity on 5/25/25, 8:41 PM
As well as been playing with creating plastic keys using a flatbed scanner with the printer.
by bitwize on 5/26/25, 4:31 AM
by bananatron on 5/26/25, 6:19 PM
by tomek_zemla on 5/25/25, 9:53 PM
by wila on 5/26/25, 12:04 PM
AntView is an ActiveX wrapper for the Microsoft WebView2 component. AntView enables programming languages that cannot use the WebView2 component directly to have a modern browser component in their applications.
You can find it here: https://antview.dev
by raindy_gordon on 5/25/25, 11:54 PM
I'm developing a small community focused on rating TV show endings. I've grown tired of investing time in series that get canceled and end on cliffhangers. Unless the show is really good, and even then, I prefer starting knowingly.
by EdA1 on 5/27/25, 10:09 PM
by burgerone on 5/26/25, 7:31 AM
by hilti on 5/25/25, 8:42 PM
1) Setup Apache https://github.com/marchildmann/IDS-Scripts
2) Setup MLX and MLX-LM Finished by tomorrow
3) Working on a micro PHP framework to instantly deploy an API, connect a database and have a basic middleware
by lylo on 5/26/25, 7:31 AM
Who needs Substack?
by naftalibeder on 5/26/25, 1:23 AM
My brother is always sending me articles to read, and I don't like reading at a computer, so I made a tool to send articles to my Kindle. It makes me happy and I use it every day, along with a small but apparently happy customer base.
by sp1982 on 5/25/25, 10:24 PM
by kimjune01 on 5/26/25, 2:12 PM
https://github.com/kimjune01/mcp-ghost
it's on pip!
by nazcan on 5/25/25, 10:21 PM
I got tired of missing deliveries, so now software answers the buzzer.
Using a mix of telephony, transcriptions, and websockets. Webserver is in C++.
by lostmsu on 5/26/25, 3:34 AM
After that's out (hopefully end of the week), gonna focus on ambient voice AI that would do turn taking (e.g. participate in conversations when it makes sense).
by gagik_co on 5/25/25, 9:54 PM
by throwaway874839 on 5/26/25, 9:47 AM
Original HN post: https://news.ycombinator.com/item?id=32551862
by jazzprogramming on 5/26/25, 6:39 AM
by sgammon on 5/25/25, 8:17 PM
by rodolphoarruda on 5/26/25, 5:22 PM
2. Productivity tools for independent sales reps that simulate a Sales supervisor/manager. Fights procrastination and lack of focus on daily activities.
by samschoice on 5/26/25, 4:01 AM
https://www.sharemydevstack.dev/
Still working on some updates to it like a search
Add your stack without going to the actual create your stack page etc.
by noahfk on 5/26/25, 10:53 AM
Now I'm updating the user manual to show all the new features!
by ainiriand on 5/27/25, 5:17 PM
by pawelkobojek on 5/26/25, 6:52 AM
by rpearcea on 6/1/25, 6:24 AM
by ilaksh on 5/26/25, 1:04 AM
by nirkalimi on 5/26/25, 5:21 PM
Its a super simple way for your community to submit and vote on the content that they'd like you to check out/react to next.
by gabriel-uribe on 5/25/25, 10:27 PM
Simply emails you the story with chinese characters, pinyin, etc based on your level and story topics of interest
by solresol on 5/26/25, 1:26 PM
https://bead-sort.arithmetic.guru/
Oh, and also a PhD on p-adic machine learning, but no one will remember that.
by p0deje on 5/25/25, 11:11 PM
Keep working on a test automation library that should allow writing browser/mobile tests easier with LLMs help, so I could focus more on testing, and less on automating.
by iddan on 5/26/25, 4:00 PM
by danielvaughn on 5/26/25, 1:17 AM
It’s still in development, but you can see a codebase for an earlier prototype here:
by XCSme on 5/26/25, 11:32 AM
by JanisIO on 5/26/25, 12:42 AM
by codr7 on 5/25/25, 9:41 PM
by ramon156 on 5/26/25, 7:33 AM
DBeaver was my go-to but the UI became too buggy for me. Alternatives all cost a buttload of money for basic features and I couldn't be bothered spending money just so I can make a backup of my DB
by rpastuszak on 5/26/25, 10:00 AM
2. Getting back to writing regularly after a health break (https://untested.sonnet.io)
by nahumfarchi on 5/26/25, 4:33 AM
Not much yet, it includes a basic half-edge implementation and a couple of small examples.
by Kuyawa on 5/26/25, 8:59 PM
You can take a look at https://my.adminix.app/demo
by hoppp on 5/28/25, 8:22 PM
Its blockchain dev using the flow blockchain's verifiable random function, basically gambling where the games are NFTs and the fees go to NFT owners.
by artrockalter on 5/27/25, 7:23 AM
- single page site for keeping track of what you're reading/watching (building for my parents who use pen and paper for this)
by lunarcave on 5/25/25, 9:17 PM
It's a Typescript library that allows you to wrangle structured outputs from LLMs and pipe them to programmatically useful control flow or structured data.
by tonyedgecombe on 5/26/25, 10:21 AM
by ema on 5/26/25, 7:37 AM
The idea is to aggregate videos by language and then sort them by difficulty.
by ALLTaken on 5/26/25, 10:31 AM
Novelty: Massively Agentic multi-modal AI Model with task-based agent synthesis and no measurable hallucinations. New consensus algorithm able to integrate empathic reasoning and emergent motives.
by gjones779 on 5/27/25, 12:25 AM
by litemn on 5/25/25, 10:25 PM
by departed on 5/29/25, 3:01 PM
by Zamaamiro on 5/25/25, 9:30 PM
The approach is to perform system scanning using a combination of LLMs and traditional algorithms to dynamically populate a Datalog knowledge base. The facts of the program are constrained to a predefined “model schema” of sorts and a predefined set of rules that encode specialized domain knowledge of how new facts can be derived from known facts.
We generate proof trees / attack graphs from the knowledge base and queries posed to it. The attack graph uses big-step semantics to plan and guide the execution flow, and the system dispatches to agents with tool use to fill in the details and implement the small-step semantics, so to speak. This may include API calls to a Metasploit Framework server or RAG over vulnerability and exploit databases.
We use Pydantic AI to constrain the LLM output to predefined schemas at each step, with a dash of fuzzy string matching and processing to enforce canonicalization of, e.g., software names and other entities.
Tl;dr: neurosymbolic AI research tool for cybersecurity analysis and pentesting.
by primaprashant on 5/26/25, 5:40 AM
by kebsup on 5/25/25, 11:00 PM
App with dynamic/flexible spaced repetition flashcards for language learning.
Recently I've added dialog & definition cards, so I can learn German from short dialogs with images and audio.
by ensemblehq on 5/26/25, 11:13 AM
by jlengrand on 5/26/25, 8:53 AM
by bl4kers on 5/25/25, 11:29 PM
by torvald on 5/26/25, 10:56 AM
by henning on 5/25/25, 9:21 PM
by hboon on 5/26/25, 12:31 AM
by breadchris on 5/26/25, 5:29 PM
by dingjiaqian on 5/27/25, 1:39 PM
by rozenmd on 5/26/25, 8:05 AM
by hemmert on 5/25/25, 9:48 PM
(I made a small newsletter sign-up form, feel free to join the wait list for betas and a free e-Book!)
by mmarian on 5/29/25, 11:02 AM
by egorbatik on 5/25/25, 10:30 PM
* We are just starting with Projects in Porto Belo - Brazil. We are adding more countries soon, but it is worth to explore the catalog.
by sch00lb0y on 5/26/25, 11:46 AM
by skoutXII on 5/26/25, 4:36 AM
by TOGoS on 5/25/25, 9:42 PM
In practice, writing journal entries about why I can't seem to get myself to make all these French cleats that I supposedly need.
Also some software stuff.
by guybedo on 5/26/25, 12:36 AM
That's what I'm working on.
by ravroid on 5/25/25, 8:24 PM
by daza on 5/25/25, 8:24 PM
by dpacmittal on 5/26/25, 4:50 AM
by nhatcher on 5/25/25, 9:33 PM
by nlh on 5/25/25, 10:21 PM
The retail site is 100% custom code built in Crystal (server) and Svelte (client). The only part that isn't running my own code is our checkout flow -- I let Shopify handle everything after "Add to Cart".
Our system backend is a separate Crystal app which handles inventory management, pricing research, and price prediction. I've developed an ML model to do price prediction and it kinda works?
What I'm actually working on: This is my full-time gig and probably 60% of my time is spent running the business (going to coin shows, buying coins, photographing new purchases, etc.) and 40% is spent writing code to make the 60% run more efficiently :). It seems I have an infinite list of things to do -- improvements to our retail site; Improvements in how to efficiently go from coin to retail listing (turns out you can send just photos of coins to Claude and with the right prompt it will actually give you a reasonably good description that doesn't sound toooooo AI slop-y); Next "big" project is adapting our ML model for paper currency. The taxonomy is similar but not the same and there's a whole world of notes out there that need to be priced.
Always happy to talk about this stuff so always feel free to email with any numismatic (or tech-numismatic) questions. noah@rarity7.com.
by wtf242 on 5/25/25, 10:55 PM
https://thegreatestbooks.org/recommendations?demo=tgb2025
warning: account required, and the full featured version where you can specify book length, include/exclude genres/subjects, etc requires a membership. if you would like to test it though just e-mail me at contact@thegreatestbooks.org and I'll mark your account as paid.
by apstyx on 5/26/25, 12:04 PM
Considering expanding the service for REST endpoints.
by flashblaze on 5/25/25, 8:33 PM
I redesigned the home page today itself. Any feedback is appreciated!
by mxrio on 5/26/25, 12:49 AM
by jfim on 5/27/25, 9:02 PM
Pretty fun project!
by jelled on 5/26/25, 12:19 AM
You can grab a TestFlight link at https://cutword.com
by stitched2gethr on 5/26/25, 12:49 AM
by wsintra2022 on 5/26/25, 1:00 AM
by martinrue on 5/26/25, 5:30 PM
by Igor_Wiwi on 5/26/25, 3:17 PM
by sameg14 on 5/26/25, 6:34 PM
by pants2 on 5/25/25, 8:51 PM
Open source Mac-native menu bar app for speech to text using GPT-4o-transcribe (current STT SOTA)
by codazoda on 5/26/25, 2:38 AM
You might have read my previous article about How To Lose Money with 25 Years of Failed Businesses.
by vicarrion on 5/26/25, 3:15 PM
by taormina on 5/25/25, 11:48 PM
Flutter + Flame + Spine + YarnSpinner. After a year of development, we’re coming up on some very fun milestones!!!
by weiwenhao on 5/26/25, 4:14 AM
by roland35 on 5/26/25, 12:31 AM
by busymom0 on 5/25/25, 11:33 PM
It's not exactly version 2.0, it's built entirely from scratch and instead of only hacker news, it can also be used for similar forum sites like Lobste.rs, Tildes, Lemmy etc. In fact, it's built in a way such that more website support can be easily added on the fly.
I had restarted this 3 times in the last 2 years. But the current code is finally coming together to be released to the public.
Currently, I already have the reader part working. So one can read posts, comments, expand collapse comments, read articles etc. I don't have the writer part working yet (voting, favoriting, commenting). I am debating whether I should just release the reader part first and then continue working on the writer part and release it as part of update. Thoughts?
by eqmvii on 5/25/25, 8:32 PM
It doesn’t feel there yet, but starting to seem some workflows could be close. And non-technical folks at business are starting to pay attention and want projects moving in those areas.
by jpcerrone on 5/26/25, 2:15 PM
by AminZamani on 5/29/25, 8:09 PM
by dirteater_ on 5/26/25, 5:27 AM
by rriley on 5/26/25, 2:28 AM
An LLM answers in-character; Try it free: https://roleplayr.ai
Would love feedback.
by cadr on 5/25/25, 10:14 PM
by aproductguy on 5/26/25, 6:36 PM
by harubi on 5/26/25, 8:55 AM
A free macOS that does TTS to any highlighted text in any app with AI voices. A Speechify alternative.
by herol3oy on 5/25/25, 8:34 PM
by bradly on 5/26/25, 12:35 AM
by strontian on 5/26/25, 4:10 AM
pastybara.com
by mark336 on 5/26/25, 12:06 AM
by robch on 5/26/25, 4:14 AM
by digest on 5/26/25, 10:30 AM
by Wdorf on 5/28/25, 5:56 AM
by Yiling-J on 5/26/25, 12:31 AM
by tomlockwood on 5/26/25, 6:14 AM
by alexpadula on 5/26/25, 3:04 PM
by alexnastase on 5/25/25, 8:52 PM
by nbbaier on 5/26/25, 2:49 AM
by nozmoking on 5/25/25, 10:22 PM
by jingntonic123 on 5/26/25, 2:22 AM
by jingntonic on 5/26/25, 2:22 AM
by metladsl on 5/26/25, 8:04 AM
by Eikon on 5/25/25, 11:19 PM
by simquat on 5/25/25, 9:41 PM
by aduermael on 5/25/25, 10:54 PM
by level09 on 5/26/25, 8:34 AM
Futuristic CMS concept with Embedded agent to replace traditional forms.
by grxar on 5/26/25, 10:53 AM
by smith-kyle on 5/26/25, 6:27 AM
You can sign up for the beta here: https://diffender.com
by detectivestory on 5/26/25, 8:58 AM
by siliconc0w on 5/25/25, 10:18 PM
by Ygg2 on 5/26/25, 7:05 AM
by LarsDu88 on 5/25/25, 10:33 PM
by badenglish on 5/26/25, 8:59 AM
Two demo are already written: explanation how to compute parallel and algorithm of parallel space emulation.
by arsalanb on 5/25/25, 10:40 PM
An AI data scientist for serious data work. Think of it like an AI native Jupyter notebook.
by Amza on 5/25/25, 10:29 PM
by TZubiri on 5/25/25, 10:52 PM
by Yoric on 5/25/25, 8:57 PM
by bobsacamano19 on 5/26/25, 4:15 AM
I've been reading the Wall Street Journal every morning this year but I'm too cheap to pay for the hefty $39/mo. subscription. My routine was: open article → hit paywall → copy URL → paste into Archive. After a few dozen times that got old.
So I built Big 5 Archive: a tiny Chrome extension that automatically redirects any link from WSJ, NYT, Washington Post, L.A. Times, or USA Today to its archived, paywall‑free copy.
I had fun making this and I hope it is helpful to the news-readers out there. Feedback welcome, happy reading!
by samirsd on 5/25/25, 10:31 PM
by ivylee on 5/27/25, 2:43 PM
If you want to try, send me an email.
by jmhmd on 5/26/25, 4:30 PM
Building mainly to power the next generation of pacsbin.com, but may offer as a standalone service as well.
by kilroy123 on 5/26/25, 2:12 AM
by reducesuffering on 5/25/25, 11:47 PM
by wtp30twice on 5/26/25, 12:01 AM
by elpakal on 6/1/25, 1:39 PM
by whytaka on 5/26/25, 6:14 AM
It's a democratic webring creator/management app where members can vote on new websites to join their ring.
by sm001 on 5/26/25, 11:23 AM
by r3tr0 on 5/30/25, 7:30 PM
you can checkout our sandbox at https://yeet.cx/play
by skeptrune on 5/26/25, 7:00 AM
by davedx on 5/26/25, 5:46 AM
by ravivooda on 5/26/25, 6:33 PM
For the tinkerers of the world: drop your tool, we handle auth, payments, usage. You get paid per run.
For the consumers: don’t like selling your data just to convert a file? You’ve found your home. No tracking. No ads. No dopamine hacks (only regular dosage).
by DigitalNoumena on 5/26/25, 8:49 AM
by jorisboris on 5/25/25, 11:26 PM
I have the impression clients like it when their code is “visual” so I’m trying to learn more of it to attract new clients
by kelsey98765431 on 5/25/25, 10:24 PM
by tehlike on 5/25/25, 9:47 PM
by jachee on 5/26/25, 2:41 AM
by ranuzz on 5/25/25, 8:27 PM
by thedangler on 5/26/25, 1:43 PM
Fantasy Apps
by maurodelazeri on 5/26/25, 2:18 AM
by steveharrison on 5/26/25, 7:00 AM
- Stores transactions locally in a SQLite database, so in theory you could build your own Front End if mine didn't do everything you wanted. Think someone who wants to do a fancy "year end summary" visualisation. And if you want to do a complex filter, can just write an SQL query to the database itself.
- Has a great UI.
- Allows you to write your own import connector and mix 'n' mash between locally-saved CSV/OFX files and APIs, rather than going down the standard route of only supporting Plaid. Current solutions leave users high and dry if the Plaid connector doesn't work well / they want to import old data (Plaid only gives the last 2 years I think).
by sbalamurugan on 5/26/25, 9:55 AM
by nickpeterson on 5/25/25, 8:26 PM
by pabna on 5/25/25, 9:18 PM
by fortmeier on 5/26/25, 12:15 PM
by bobofee on 5/26/25, 2:43 AM
by nandosobral03 on 5/26/25, 12:08 PM
by reverseblade2 on 5/26/25, 1:54 AM
by transformi on 5/25/25, 9:45 PM
by Zaloog on 5/25/25, 9:27 PM
by windowshopping on 5/25/25, 8:29 PM
by yamapikarya on 5/26/25, 2:11 AM
building a prometheus exporter to monitoring like temperature, cpu, ram and building simple html monitoring tool to expose those metrics.
by lonh on 5/26/25, 8:48 AM
by Toks on 5/26/25, 12:08 AM
by adityaathalye on 5/26/25, 5:08 AM
... for a little SaaS app, because my friend-and-sponsor hasn't given me a deadline, nor sent irate letters asking for results.
Think of it like the poor man's XTDB + Datomic, if you like. Readings and conversations have taken me to deep questions explored in works by Snodgrass [0], Jensen[1], and Jepsen [2]. Shout out to the XTDB folk for their feedback and suggestions. This is why I hang in the Clojure community... it's full of people doing rad stuff who are also unwaveringly kind, helpful, and generous with their time.
---------------------
The pre-alpha version of "Project Writing for Nerds" [3]
A cohort-based workshop to help fellow gentlenerds "spool brain to disk" with as little friction as possible.
---------------------
[0] Developing Time-oriented Database Applications in SQl https://www2.cs.arizona.edu/~rts/tdbbook.pdf
[1] Temporal Database Management https://people.cs.aau.dk/~csj/Thesis/
[2] Jepsen's helpful little notes on Consistency https://jepsen.io/consistency
[3] https://www.evalapply.org/index.html#writing-for-nerds
(edit: fix formatting)
by curlcntr on 5/26/25, 1:01 AM
by solomonb on 5/26/25, 6:42 AM
by bbx on 5/25/25, 9:28 PM
by muhammadusman on 5/25/25, 8:23 PM
by gnuser on 5/26/25, 1:53 PM
by meowzarella on 5/25/25, 10:59 PM
it's a tarpit idea that a lot of users and investors like to shit on so i decided to just build something that i like myself.
by rurban on 5/26/25, 10:19 AM
by GMoromisato on 5/25/25, 10:26 PM
It's basically a full-stack web platform written entirely from scratch. One of these days I'll write about it and get yelled at for reinventing the wheel.
But I'm using it internally and for my biotech clients and I'm still excited about it.
by elpakal on 5/25/25, 9:45 PM
by sumeruchat on 5/26/25, 8:26 AM
I got everything done technically but marketing is challenging
by ciguy on 5/26/25, 12:31 AM
by bobowzki on 5/26/25, 7:56 PM
by arionhardison on 5/26/25, 1:38 AM
by SpecialistK on 5/26/25, 6:33 PM
by peab on 5/25/25, 10:40 PM
A web app for people to get tarot readings, and create their own tarot cards using AI. I'm enjoying working on this because I'm using as an opportunity to learn parts of the stack that I didn't usually do at my day job - frontend, design and marketing (my career has focused more on the backend).
by viksit on 5/25/25, 11:34 PM
rather than manually write prompts for llms (which is like hand coding byte code for cpus), declare a structure and instructions and let the system do prompt writing for you.
it also exposes an optimizer which can do sophisticated prompt learning for tasks.
github.com/viksit/selvedge
by brynet on 5/25/25, 10:25 PM
https://brynet.ca/wallofpizza.html
But seriously, I'm looking for "no-strings" sponsors, if any companies (or individuals) would like to help support me so that I can focus on open source full time. Feel free to email me: https://brynet.ca/
by trumbitta2 on 5/30/25, 10:19 AM
by MarceColl on 5/26/25, 1:09 PM
by prmph on 5/25/25, 8:47 PM
- Manages the entire range of personal (and maybe business) information/content: Documents, Media, Messages (email, instant, etc.), Contacts, Bookmarks, Calendar, etc.
- Is tag based, so that where to put and find content is easy to answer. Think of a set of flat folders, on one or more devices, within which the files are stored with tags attached. Since people often find navigating/browsing files more natural than searching, virtual folders will be dynamically generated to guide navigation. Also, entire folders can be treated as atomic, and tagged/managed as one object (useful for repositories & projects). And, heuristics (and maybe AI) will be used to automatically tag files when they are imported into the tool, greatly reducing the tedium of adding tags.
- Is file based, so that all information is physically stored as individual files. This allows information to be more easily managed on a physical level: moved around, backed up, exported/imported, searched, navigated, etc. So in addition to docs, each email/instant message, contact, scheduled task/event, bookmark, etc. would ultimately be stored as a file, unlocking all the things you can do with files.
- Has a local web-based UI launched from a local agent, so actual file content does not usually need to move across the network and stays local, and the tool is also easily multi-platform, with consistent UI irrespective of platform.
- Provides a cloud web UI as well, that communicates with content devices through the local agent, so that content stored across multiple devices can be managed in one central location, even without direct access to those devices, team/org features can be provided. However, file content still stays local, except when shared.
- Provides tools for exporting data as file from the data islands of various apps and service, and backing up as files to cloud storage services.
My vision is a situation where I am in charge of my own data irrespective of whatever device, app, or service I use, can ensure that it is always available and will not be lost, and that I can easily navigate and search through it all to find whatever I want, no matter how scattered and massive it is.
[1] Here are some of my issues with personal information management affordances of current tech, which is driving me to work on a solution:
- Our data is too bound to device and vendor islands. Can't easily move my information across Apple/Google/Whatsapp, etc accounts. Can't easily merge and de-duplicate either. I almost always somehow lose data whenever I have to move to a new phone, etc.
- Hard to own your data on many services: Discord, Slack, etc. Can't easily export, search.
- Hard to have a 360 overview and handle on all your data assets and query them in a consistent manner.
- Files as a unit of information storage and management is very ergonomic; we shouldn't allow that concept to be buried by vendors for their own gain.
by dabei on 5/26/25, 2:06 AM
by nhanotia on 5/26/25, 3:56 AM
by nhanotia on 5/26/25, 3:56 AM
by czottmann on 5/26/25, 8:59 PM
Now I struggle with a marketing strategy, quite a bit. But hey, I was told it's part of the job ;)
by dom96 on 5/26/25, 9:41 AM
I am also integrating it with ATProto.
Current MVP supports search only: https://imaginarium.picheta.me/
by TMEHpodcast on 5/25/25, 11:09 PM
by RobRivera on 5/25/25, 11:07 PM
by contingencies on 5/25/25, 8:42 PM
by semessier on 5/25/25, 10:15 PM
by personjerry on 5/26/25, 5:55 AM
by aurelz993 on 5/29/25, 8:39 PM
by hiAndrewQuinn on 5/26/25, 5:44 AM
The most exciting thing is I'm working on a way to pre-bake one's WiFi information and an SSH key into the piCore Tiny Core Linux image.
piCore is sort of the holy grail for a lot of people who just want to stick a tiny personal-user server onto a Raspberry Pi and then not have to worry about the SD card burning out on them, because the entire system runs in RAM after first boot anyway - the only thing standing in its way, methinks, is that it requires you to physically plug in a keyboard and a monitor to get it up and running on the first boot.
I think that's dumb. A Raspberry Pi should be a thing you can plug into a wall socket in your home and then work entirely via SSH with. I do a lot of work with Linux images and virtualization like this at my day job anyway, with more scrutiny and regulatory compliance in place - if you can imagine what "embedded devops" might look like, that's a good way to describe it. So I'm a surprisingly good fit by now for this kind of thing.
===
Currently I'm neck deep in some Fish shell scripting to automate writing posts for https://andrew-quinn.me/ , similar to what I already have for https://hiandrewquinn.github.io/til-site/ . With the latter, I just type in `til` at a terminal, and I
1. Get asked for a title;
2. Get a fzf multi-select list of all previous tags used on the blog (this ensures I actually continue to use the same handful of tags, which makes the blog much more fun to dive down); and
3. Get a vim window open to actually crank out the blog itself.
aqdm doesn't have tags, but it is built on a very similar Hugo base. Honestly the hardest thing about writing this script is simply wrestling with the fish shell programming language. I should really sit down and learn it properly one day...
===
We changed our modem recently, which broke the cronslave Raspberry Pi I had powering https://hiandrewquinn.github.io/selkouutiset-archive/. That led me to spend the weekend finally moving its codebase off of Raspbian and onto a fully RAM-based Tiny Core Linux.
Running the OS entirely out of RAM for a Raspberry Pi increases its longevity by at least an order of magnitude, because you only ever read from the SD card on boot and on backups. If you configure the server such that everything is ephemeral anyway, you'll basically never see that little green "Writing" LED blink. That's how I got interested in the aforementioned prebaked piCore project!
===
I collected my smattering of Finnish language learning software projects all in one place at http://finbug.xyz/ . I recently added the ability to mark and unmark words to my Finnish pocket dictionary https://github.com/hiAndrewQuinn/tsk - that makes it super easy to save words during a reading session and then import them into Anki later.
Of more general interest to people might be https://github.com/hiAndrewQuinn/audio2anki , which takes a YouTube URL, runs it through Whisper for an automatic transcription, then slices it up sentence-per-Whisperified-sentence and creates Anki cards out of each transcription for spaced repetition listening practice.
I made this for Finnish originally, but it fills a niche I was surprised to see had no really good locally-runnable FOSS options. There are some startups hawking a similar slice-and-dice service which I'm sure is much higher quality. But I figure the most likely people to get a lot out of a program like this are high school students, who generally don't have credit cards they can buy SaaS with - they need something they can dump their hours into, to get to work, instead of their dollar bills to get something that already works.
by wenc on 5/26/25, 2:48 AM
Because it's hard to remember how to handwrite complex kanji (many people have character amnesia in real life due to smartphone usage), I casually wondered if it was possible create a Japanese orthography that was: (1) easily scannable (which rules out hiragana); (2) disambiguated words without Kanji; (3) still relatively compact? (which hiragana is not).
I figured a good substrate to start from was romaji + a new emoji system.
You know how people think LLMs can't invent things? o3 just invented this system whose goal is to maximally disambiguates homophonic Japanese words (performing the same semantic compression role that Kanji does today). This is the first iteration. After each romaji noun, it tags it with a geometric shape. These are the 30 tags o3 came up with (based on homophones requiring disambiguation):
| Diacritic | ◐ living | ▢ built / object | ⬢ nature | ◊ abstract | ⟐ action / event |
| ------------- | ---------------------- | ---------------------- | ----------------------- | ---------------------------- | --------------------------- |
| • top dot | ◐• adult / main person | ▢• building / place | ⬢• plant / flora | ◊• idea / thought | ⟐• movement / travel |
| – bottom bar | ◐– child / minor | ▢– tool / instrument | ⬢– water / liquid | ◊– emotion / feeling | ⟐– communication / speech |
| + right plus | ◐+ group / collective | ▢+ vehicle / transport | ⬢+ weather / sky | ◊+ social tie / relationship | ⟐+ creation / production |
| × left cross | ×◐ animal (non-human) | ×▢ food / consumable | ×⬢ mineral / material | ×◊ value / moral | ×⟐ perception / sense |
| | center bar | ◐| deity / honorific | ▢| document / media | ⬢| terrain / landscape | ◊| state / condition | ⟐| change / transformation |
| ‿ bottom arc | ◐‿ anatomy / body part | ▢‿ container / vessel | ⬢‿ energy / fire | ◊‿ knowledge / data | ⟐‿ rest / passive state |
現代における日本語の一般的な表記法は漢字仮名交じり文であり、漢字と平仮名（昔の法令などでは片仮名）を交えて表記する。漢字は実質的な意味を表す語に使われ、平仮名は主に活用語尾[注 1]や助詞に使われる。朝鮮語などでは漢字が主として字音語にしか使われないのに対し、日本語では和語にも使われ、外来語を除いてほとんどの語に使うことができる。煙草（タバコ）や合羽（カッパ）、珈琲（コーヒー）など大航海時代以降にヨーロッパから入った語彙には、外来語であるにもかかわらず漢字が使われるものがある。
gendai◊| ni okeru nihongo◊‿ no ippan-teki na hyōkihō◊‿ wa kanji▢| kana▢| majiri bun▢| de ari, kanji▢| to hiragana▢| (mukashi no hōrei▢| nado de wa katakana▢|) o majete hyōki◊‿ suru. kanji▢| wa jisshitsu-teki na imi◊• o arawasu go◊‿ ni tsukaware, hiragana▢| wa omo ni katsuyō-gobi◊‿ ya joshi◊‿ ni tsukawareru. chōsengo◊‿ nado de wa kanji▢| ga shutoshite ji-on go◊‿ ni shika tsukawarenai no ni taishi, nihongo◊‿ de wa wago◊‿ ni mo tsukaware, gairaigo◊‿ o nozoite hotondo no go◊‿ ni tsukau koto◊‿ ga dekiru. tabako⬢• (tabako) ya kappa▢– (kappa), kōhī×▢ (kōhī) nado dai-kōkai-jidai◊| ikō ni Yōroppa⬢| kara haitta goi◊‿ ni wa, gairaigo◊‿ de aru ni mo kakawarazu kanji▢| ga tsukawareru mono◊‿ ga aru.
(of course, it's incorrect in parts... because LLM)
But the idea has legs. It will probably not replace Kanji due to cultural inertia, but this could a kind of shorthand especially for handwriting.
I'm pretty impressed! o3 just invented something new. It combined romaji and a tag system that it hallucinated to design a new Japanese orthography. As far as I can tell (I could be wrong), something of this nature has not been done before.
by flatthibaut on 6/1/25, 7:20 PM
I've decided to release an MVP so I can gauge interest, and after ~8 months of deployment and 2 releases there has been some great feedback. It's definitely a much much bigger and more complex project than I originally anticipated though!
At the moment I paused new sign ups as it's not scaling up nicely, so I'm now refactoring the services to use async / message queue for the heavy Python tasks using Celery / RabbitMQ.
It was quite difficult to get the product in front of people that might interested and even though I found enough users to test and give feedback, it's been tough to find them. So my strategy now is to try and release blog posts on my architecture, approaches, pros/cons, etc, and build in public regularly. This takes away time from the actual development but hopefully I'll have a big enough community that I can tap into for testers when the new release is out.
I'm not posting a link to the project or even mention the name as I got a shadow-ban in Reddit when I posted a summary to my MVP post-mortem (no link) and now I'm terrified!
I'd appreciate any suggestions / different approaches you might have in terms of reaching more users!