
Write a context-window strategy for your codebase
You'll have a one-page plan describing which parts of your system each engineer should feed to AI tools, so everyone maximizes useful context without hitting token limits.
What you'll make
AI Context Strategy
- Context Zones: which code sections each role should include
- Preparation Rules: how to clean and focus code before feeding to AI
- Quality Signals: how to recognize when context choices are working vs failing
- Team Standards: shared practices for consistent context across tools
Write a one-page strategy document with specific code paths and team role assignments
The 3 parts
Each part ends with a check you can see.
Analyze Your Context Workload
Map which code areas each team member works in and identify the complexity patterns that determine context needs
Done when
You have a workload map showing each team member's primary code areas with complexity ratings and context decision rules for each zone
Test Context Choices on Real Tasks
Run different context preparation approaches on actual coding tasks and measure which produce better AI output
Done when
You have tested at least 3 different context strategies on the same coding task and can explain which worked best and why
Design Team Adoption Strategy
Decide which context rules your team will actually follow and design enforcement mechanisms that fit your workflow
Done when
You have specific team adoption decisions with enforcement mechanisms and a complete context strategy document ready for implementation
Capstone: you put together your AI Context Strategy.
Where this helps
An elective: a single lesson that sits outside rtwo's job certificates and doesn't count toward one.
About this lesson
What you'll be able to do
- Create a one-page context strategy that assigns specific code sections to each team member for AI tools
- Run your strategy on a real feature branch and identify which context choices actually improved AI output
- Have a tested set of context preparation rules your team can use consistently across different AI coding tools
Who it's for
- Development team leads who want to standardize how their team uses AI coding tools
- Senior developers tired of teammates feeding random code to AI and getting generic output
- Teams that have multiple people using AI for code but no shared strategy for what context to include
What it won't cover
- Choosing between different AI coding tools or setting up API access
- Teaching basic prompting techniques or AI fundamentals
- Code architecture decisions or refactoring strategies unrelated to AI context
Skills you'll practise
- Context boundary analysis for large codebases
- AI output quality assessment in code review
- Team workflow standardization
- Context preparation methodology