Revolut runs coding agents for 2,000 engineers at 1/10th cost with open models
zainhas · x · 2026-11-09
Revolut's AI Engineer New York talk: serving coding agents to 2,000 engineers using open models (GLM 5.3 on 36x B200), delivering products 70% faster at 1/10th the bill while staying compliant. Key lessons: tune hyperparameters before scaling (80% of gains came from tuning); watch TTFT as Slack message volume grows; token usage is a non-issue.
More from coding & agent
- 'Headless browser' dubbed the creepiest tech term as agents inspect 120 cards — yuwen_lu_ · 2026-11-09
- edge-arm-bridge: mirror Edge profiles to native arm64 Chromium so agents run at native speed — unixterminal · 2026-11-09
- Agent Buddy: an ESP32 animated companion that shows your AI coding agents' live status — DanWahlin · 2026-11-09
- Open-source Skill recreates the viral "flat illustration opens into 3D" dolly-zoom video style — FinanceYF5 · 2026-11-09
- levelsio feeds 58 minutes of DJI clips to Claude Code, gets a 10-minute vlog back — alvelda · 2026-11-09
- OpenAI's LifeVal maps chore demand from 3,995-person survey — but benchmarks aren't proof of paying customers — krishnan · 2026-11-09