AI-assisted code changes
Specify and review bounded changes while keeping responsibility for the result.
Start with a change small enough to explain
A customer sets a quantity to zero, but the application restores the old value. You will turn that symptom into a precise contract, inspect a proposed change, and explain what evidence supports the repair. AI can draft code, but it cannot decide the product’s missing rules for you.
Begin with the change loop, then use the zero-quantity exercise. You finish with a reviewable patch, a concrete input/output example, and a record of assumptions. No cloud account is needed.
Parts group related chapters. Each lesson has a chapter.lesson address, such as 4.07. Open a title below, or use Next to follow the reading sequence. Within a lesson, On this page lists its sections.
- 3.01
Make code changes reviewable and recoverable
Concepts and examples
- 3.02
Specify and verify an AI-generated change
Concepts and examples
- 3.03
Repair a quantity update that loses zero
Concepts and examples
- 3.04
Estimate request rates, storage, latency and availability
Concepts and examples
- 3.05
- 3.06
- 3.07
- 3.08
- 3.09
- 3.10
- 3.11
- 3.12
- 3.13
- 3.14