Complete this onboarding to understand the tutorial codebase structure, code style, and quality standards.
Before starting, understand these non-negotiable principles:
-
Avoid AI slop at any cost
- Humans can instantly detect low-quality AI-slop
- Once detected, everything else sounds equally horrible
- A tutorial cannot be "throw a prompt into Claude Code and copy-paste the
result"
- It's ok to use AI, but the output needs to be higher than using only AI or using only human
- Your goal is to read information and synthesize that for somebody else
to reduce the time it takes for them to learn by 10x
- If it takes 5 mins to generate and 60 mins to generate
- Humans can instantly detect low-quality AI-slop
-
Use a professional tone
- Write as if explaining to a peer, not a student or a buddy
-
Focus only on examples without repeating content
- Everything should be stated only once and in the right place
- Eliminate redundancy between sections
-
Maintain high quality standards
- Every sentence should serve a purpose
- Each example should illustrate a concept clearly
Create an issue called "Clean up tutorial XYZ" with the following action items.
- Read
class_project/project_template/README.md- Is there anything unclear, incorrect, or that can be improved?
- Carefully examine all code in
class_project/project_template- Make sure you understand every file and module perfectly
- Note the structure, naming conventions, and patterns used
-
Become familiar with the Claude Skills in
helpers_root/.claude/skills- Skills are living documentation of how we maintain and create code
-
Explore the available skills using these commands:
> md skill describe coding > md skill describe notebooks > md skill describe testing > md skill describe X_in_60_minutes
-
Review key skills relevant to tutorials:
coding.format_rules: Python code conventionstesting.format_rules: Unit test patternsnotebook.format_rules: Jupyter notebook structuremarkdown.format_rules: Documentation style- E.g.,
> md skill edit coding.format_rules
-
Is there anything unclear or that could be improved in the skills?
- Read 2-3 tutorials
- Tutorials that are closer to the standards are:
- Autogen
- Internalize its structure, style, and approach
- Note patterns, structure, and what makes them effective
- Identify what works well and what could be improved
- Tutorials that are closer to the standards are:
- Review the test structure, specifically
class_project/project_template/test/test_docker_all.py- Understand what tests each tutorial should include
- Ensure your tutorial will have comprehensive test coverage