plan.md

1. Background
   • Operating systems (Don't change, record)
   • Graphs (Don't change much, record)
   • Filesystem (Very short, Linux/MacOS vs. Windows)
   • Floating point numbers (Find good resource)
2. Tools
   • Shell: Basics / background (Short, mostly old material)
   • Shell: Navigation (New(ish))
   • Shell: Installation of Mamba and course environment (Video, text)
   • Installation: Git, VS Code (only text / links)
   • Editors, comparison IDEs (short)
     • VS Code
     • Stata / Matlab / RStudio
     • In the end, VS Code encompasses most features of IDE + same for all languages
   • Using LLMs effectively [long run]
   • Using copilot effectively [long run]
   • Communicating effectively (emails vs zoom vs in-person vs zulip) (Video)
   • Licenses (Look up stuff)
3. Git and GitHub
   • Motivation and terminology (record anew)
   • Committing and Diffing (Look up stuff)
   • Pushing and Pulling (Look up stuff)
   • Merging (Look up stuff)
   • Troubleshooting (Look up stuff)
   • Collaboration and Teamwork (Look up stuff)
4. Python basics
   • Assigning variables, built-in scalar types (record anew)
   • Built-in container types (record anew)
   • Reading tracebacks (record anew)
   • Importing, Modules, Namespaces
   • Filesystem in Python: The pathlib library (record anew)
   • Executing Python code: .py files (record anew)
   • Executing Python code: Notebooks (record anew)
   • Executing Python code: Debuggers, pytest, pytask (record anew)
   • Decorators (record anew, further material: Harrison)
   • Custom data containers
   • Reading object-oriented code [long run]
   • More on f-strings (could also do as an exercise)
5. Best practices (any language)
   • Avoid code duplication (iteration, functions) (record anew)
     • Objective: Understand Concept. Do stuff, see patterns that come out
   • Pure and testable functions (record anew)
   • Comments should be function names (record anew)
   • Avoid nested if conditions (record anew)
   • Naming conventions / formatting / pre-commits (record anew)
   • Choosing container types (list / dict / custom class) (record anew)
   • dags [long run] - pytrees [long run] - Don't be too smart (link) (set up examples, make scorecards / notebooks out of those, record short video)
   • When to code up something yourself and when to use a library [long run], (record screencast with recipes)
6. Ensure Correctness - Testing
   • Debugging
   • Error handling
7. Workflow management
   • Definition of "reproducible" - Pytask overview
   • Tiny Pytask example
   • Generating many tasks at once
   • Templates
8. Scientific computing
   • Numpy/scipy
   • Optimisation
   • Estimagic
   • Speedup
9. Data Science
   • Data Management
   • Plotting
   • Machine learning with Scikit-learn
   • (Econometrics with ...)
10. Text data and texts
    • Encodings
    • Regexes [long run]
    • Simple Markdown
    • MystMD
    • Slidev
    • LaTeX
    • Detour: How to read a paper
    • Detour: How to structure a paper
    • Tables
    • Figures