Essays by Ry Walker

Code production is no longer the constraint. Deploy pipelines, feature flags, and code review are. The new bottleneck is taste, and taste does not scale.

Autocomplete, interactive agents, background agents. The bottleneck has moved from generating code to reviewing it, and almost no one is building for the new constraint.

No single number captures developer performance. The honest answer is a portfolio of imperfect signals, used to develop people rather than rank them.

Any developer performance metric can be gamed. AI tools just give us new things to measure — and new ways to get it wrong.

The interesting use of AI coding tool data isn't ranking. It's understanding how your best engineers actually work — and helping the rest of the team catch up.

AI assistants are generating a granular log of every prompt, accept, reject, and iteration. That's new data — and a new way to measure the wrong thing.

Sales has revenue. Engineering has hand-waving. The asymmetry in how we measure go-to-market versus R&D is a real problem, not a feature.