A setup guide with twenty steps is a list of ways to fail. I've started treating 'clone, run one command, see the app' as a feature of every codebase. Here's what goes into that command, and the design choices that make it possible.
Delivery · Infrastructure · Codebases6 min read
Why I still build products against a one-day clock, and the habits that make it work: cutting scope hard, using the same stack every time, letting agents do the heavy lifting, leaning on a design system, and agreeing on what 'done' means before I start.
Product · Prototyping · Agents · Design systems8 min read
In my agent rig, prompts, routes, skills and even fine-tuned models all start in shadow. Promotion is a separate, evidence-backed step that a person takes. Here's the lifecycle, and why it matters more as the number of moving parts grows.
Agents · AI · Local LLMs · Delivery6 min read
A rock tumbler doesn't make a stone smooth in one pass. It uses coarse grit, then finer grit, then polish, and checks between stages. That's how I build agent loops in Starship: bounded retries, deterministic gates, and lessons that are only allowed to hurt once.
Agents · AI · Architecture · Delivery7 min read
How Starship's coding agents go from a planned task to a merged pull request without a human driving each step, and the gates, retries and postmortems that make that safe to leave running.
AI · Agents · Architecture · Delivery6 min read
Performance work on a data pipeline is only safe if it can't change the answers. Here's how golden files and output-equivalence gates let me speed up a clinical pipeline without moving a single score.
Data · Architecture · Delivery · Healthcare6 min read