
Map your actual coding patterns before letting AI generate architecture
You'll have a documented map of your system's real constraints and a decision log showing where you overrode AI suggestions—and why.
What you'll make
Architecture Reality Check
- System Constraints Map: performance, legacy, team
- AI Override Log: decisions where local knowledge trumped AI suggestions
- Workflow Documentation: your repeatable process for AI-assisted architecture
The 3 parts
Each part ends with a check you can see.
Document Your System Reality
Map the constraints AI can't see that shape every architectural decision in your codebase
Done when
You have a documented constraint map covering performance bottlenecks, legacy dependencies, and team capabilities that AI couldn't discover
Test AI Against Real Decisions
Take a recent architectural decision and see where AI's suggestions miss your system's reality
Done when
You have tested AI's architectural advice against a real decision from your codebase and documented 2+ places where local knowledge trumped AI suggestions
Build Your AI-Assisted Process
Design a workflow where you map constraints first, then use AI to explore options within those boundaries
Done when
You have a documented 3-step process for using AI on architectural decisions that starts with constraint mapping and ends with justified overrides
Capstone: you put together your Architecture Reality Check.
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
- Map your codebase's actual constraints and patterns before AI generates anything
- Override AI architectural suggestions with documented reasoning based on your system's reality
- Develop a workflow where you lead architectural decisions and AI fills implementation gaps
Who it's for
- Developers who've been letting AI drive too many architectural decisions
- Teams experiencing code quality drift from over-relying on AI suggestions
- Engineers who want to use AI as a tool without losing their architectural judgment
What it won't cover
- How to write better prompts for AI coding tools
- Evaluating different AI coding platforms
- Building AI agents or automated development workflows
Skills you'll practise
- System constraint analysis
- Architectural pattern documentation
- AI output evaluation against real constraints
- Decision justification under uncertainty