
Record a documentation-grilling workflow demo
You'll have a 5–8 minute recorded walkthrough showing your team how to use AI to extract and organize implicit knowledge from existing code, turning it into structured decision records.
What you'll make
Documentation-Grilling Workflow Demo
- Open: the problem and what changes with long-context AI
- Setup: preparing a codebase section for interrogation
- Demo: the actual grilling workflow in real-time
- Extract: turning AI output into structured decision records
- Close: when to use this vs when to skip it
Upload a 5-8 minute screen recording showing the workflow on real code from your team
The 2 parts
Each part ends with a check you can see.
Validate AI Against Known Decisions
Run AI extraction on code where you already know the architectural reasoning to test what it gets right vs what it misses
Done when
A scored comparison showing AI accuracy on 3 known decisions, with specific criteria for when extraction is trustworthy vs when human validation is required
Record Live Extraction with Real-Time Validation
Screen record yourself running the workflow on fresh code, showing both the extraction process and how you validate AI output using your tested criteria
Done when
A 5-8 minute recording that demonstrates the complete workflow with live narration explaining when to trust AI output vs when to add human context
Capstone: you put together your Documentation-Grilling Workflow Demo.
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
- Record a 5-8 minute walkthrough demo that your team can actually follow to extract decision logic from legacy code
- Build a repeatable workflow for turning undocumented code choices into structured decision records
- Demonstrate how long-context AI changes the economics of technical documentation
Who it's for
- Development team leads who need to onboard new developers faster
- Senior engineers tired of explaining the same architectural decisions repeatedly
- Teams with legacy codebases where the 'why' behind code choices lives only in people's heads
What it won't cover
- Setting up specific AI tools or API keys
- General prompt engineering techniques
- Code refactoring or rewriting — this is about documentation extraction, not code changes
Skills you'll practise
- Live demo construction
- Decision validation against team scenarios
- Long-context workflow demonstration
- AI output verification for architectural reasoning