EN·ES·PT
field notes — proof-first engineering

Working notes, written between commits.

Short, specific, and grounded in systems I actually shipped. Longer-form writing lives at sageafterdark.com.

Eighteen agents audited a live curriculum. One lesson was teaching a false error. I pointed an audit fleet at 34 published programming lessons: six auditors executing every code claim, six rewriters, six independent verifiers. 73 defects, 17 critical or high — every one anchored to a verbatim quote. Thirteen minutes of wall clock. 5 min · read → Five pages, five agents, one design file — zero merge conflicts A parallel agent fleet implemented five production marketing pages 1:1 from design files in about five minutes of wall clock. The interesting part isn't the speed — it's the three rules that made the output shippable. 4 min · read → The day my own quality gate blocked me At 11:39 my proof loop refused the PROVEN verdict — 15 high/critical CVEs had drifted into production deps. By 14:35 it was green. Both runs are published verbatim. 4 min · read → LLM regression testing with Promptfoo in CI: the minimum viable gate 30 golden traces, one judge, one failing exit code. The smallest setup that stops a bad prompt change from reaching production — runnable this afternoon. 6 min · read → No fake green: what a proof ledger taught me about AI agents Agents will happily report success they never earned. The fix isn’t a better prompt — it’s an evidence gate the agent physically cannot talk its way past. 6 min · read → Your LLM feature needs a regression suite more than a better prompt Prompt tweaks feel like progress because nobody is measuring. Golden traces + LLM-as-judge in CI turn "it seems better" into a diff you can gate on. 5 min · read → Automation that never talks to your customer The best-converting AI workflow I’ve shipped sends zero AI-written messages. Where to put the human approval point, and why it beats full autonomy. 4 min · read →

free artifact: the LLM pre-launch eval checklist →