@elliotgreen
Startup CTO. We're 4 engineers pretending to be 12, which is the whole point.
Reputation is earned per category — these are narrow, evidence-backed badges, not one karma number.
Fifteen lines, ~180 tokens, and it outperforms the 400-line files it replaced. Everything an agent needs on every task; nothing it needs only sometimes. The controversial part is what's missing.
We run both: Codex for new packages, Claude Code for anything touching existing code. The comparison frames it as either/or but the strength…
on Codex vs Claude Code for large monorepos · 3h ago
Just wired this into a GitHub Action triggered on codex/ branches. Review lands as a PR comment ~3 min after push. Happy to write it up if t…
on Use Claude to review Codex output, step by step · 3h ago
Data point from the other end: we went from four agents back down to two, deliberately. The 2.2x throughput at 4x coordination number matche…
on One agent or many: how do you split the work? · 3d ago
Small-team translation: we can't afford the automation harness, but the core finding — deploy-correlation is the high-value low-risk slice —…
on Agents on call: letting Claude triage pages · 5d ago
That audit is the whole method, honestly. The file isn't small because small is a virtue — it's small because it's evidence-based. Keep what…
on The minimal AGENTS.md that actually works · 3w ago
Data point: same task (add invoice proration) run scoped vs unscoped. Unscoped: 34 tool calls before the first edit, and it 'improved' an un…
on Scope Claude Code to one package in a monorepo · 3w ago