How 94 Internet-Famous Engineers Actually Use Coding Agents: Only 8 Read Every Line
virpo · hn · 2026-10-13
A developer building a code review tool studied the repos, blogs, talks and tweets of 94 well-known engineers, creating a card for each showing how they really work with coding agents — their agents, models, harnesses, rules, skills, hooks, and how they review the output. Every claim links to its source.
Key findings:
- Only 8 of 94 still read every line their agent writes
- 21 more read the riskiest parts and skim the rest
- Roughly 70% don't do line-by-line review
You can generate your own card and get suggestions on which engineer's tricks to adapt; corrections are accepted via PR.
More from coding & agent
- Cognition open-sources Devin agent plugin to patch the day's CVE wave — Scobleizer · 2026-10-23
- Fine-tuning models solo with a Grok bot and rented GPUs — Rasmic · 2026-10-23
- One prompt found pre-auth PHP RCEs: agent-guided vuln hunting timeline — chrisrohlf · 2026-10-23
- AutoRouter launches beta: auto-routes each coding task across Claude, GPT and other models to cut token spend — floguo · 2026-10-23
- How Speak uses GPT-Live and a Decisions API for ultra-responsive live AI language tutoring — pbbakkum · 2026-10-23
- Google Live API adds decisions model triggers, faster than function calls for voice agents — pbbakkum · 2026-10-23