by brauhaus on 8/2/22, 3:26 PM with 0 comments
Our work is mostly with data and our new hires are usually DS fresh out of college. They know their stuff when it comes to using Python and Spark to create ML models, but they lack experience in how to work in teams. So I listed the following topics I find vital for them to learn.
- Git Branching Strategies - Clean Code - Design Patterns - Testing - MLflow - CI/CD - IDE configuration (linters, useful Text Editor add-ons) - Code documentation - Dealing with business requirements in code - Publishing packages at private repos - Code structure (cookiecutters, package managers) - Code smells - Antipatterns - How to code review effectively - Debugging
My challenge is that because the company subscribes to Coursera - and because they want the course to contain scored tests - they want me to select courses from there. I argued that it doesn't make sense to select university courses to fill a gap many universities don't cover (or didn't in my case). Plus their syllabuses are usually crap: how can I know, for instance, that when a course says it covers "OOP in Python", the Gang of Four design patterns will also be taught? Other than wasting hours going through courses by myself, that is.
So
What I needed here are suggestions of other resources I could try to lobby my superiors - from great Udemy courses or Youtube channels.