Offline replay of 3,300 production decisions validated Jev routing in an afternoon for 20 cents
mikegiannulis · x · 2026-09-18
The author's company (AI-assisted book writing) replayed 3,300 real production decisions — source finding, message routing, "I already sent you that" detection — through Jev offline, comparing against past Claude decisions and human labels or deterministic ground truth, with pass/fail rules set before seeing results. Total cost: an afternoon and 20 cents, no live customer exposure.
Related event: Jev Outputs Structured Decisions Directly at $0.042/M Tokens(2 posts)→
More from coding & agent
- 4 deployment strategies explained via 4 visuals: feature toggle, blue-green, canary — _jaydeepkarale · 2026-09-18
- ServerKit: open-source server panel for apps, DBs and Docker hits 1.3k stars — tom_doerr · 2026-09-18
- Gary Marcus: Agents hold production credentials in a security gap nobody owns — GaryMarcus · 2026-09-18
- Amp's design philosophy: simple primitives, no tricks, let models improve it — HankYeomans · 2026-09-18
- You.com wraps AI Agentic Hackathon with self-improving agents challenge — PolarBearby · 2026-09-18
- Muse for Mac Launches With Cross-App Access; User Wires Up Perplexity Deep Research API — ChrisUniverse · 2026-09-18