How I Use AI to Design and Review Complex Changes
A practical account of using codebase research, an RFC, independent model critique, and a reviewed PR stack to design and deliver a complex change without giving up human ownership.
Practical notes on product engineering, frontend systems, prototyping, and human-reviewed AI development.
A practical account of using codebase research, an RFC, independent model critique, and a reviewed PR stack to design and deliver a complex change without giving up human ownership.
The best prototype does more than look convincing. It helps customers, product, design, and engineering make a better decision about what to build—and what not to.
A first production AI feature taught me to separate broad reasoning from smaller, checkable tasks—and to design the orchestration around what users could inspect and refine.
A testing platform should make realistic tests the easy path. Here is how I approached reusable context, network boundaries, authorization, and per-test variation for a growing frontend team.