NousResearch ran 1,393 subagents for 19 hours to clean 1M lines of Python, cutting codebase 34.4% and saving ~$2M
max_paperclips · x · 2026-09-17
NousResearch published a blog post detailing a large-scale agent run: on September 2 they tasked Hermes Agent with cleaning up about a million lines of Python. Using 1,393 subagents over 19 hours, the agent shrank the codebase by 34.4%, saving roughly $2M in engineering hours.
Related event: Nous Refactors Million-Line Codebase With 1393 Sub-agents(4 posts)→
More from coding & agent
- He runs his entire company on OpenClaw: an early look at agent-native operations — heyneighbor · 2026-09-17
- Nvidia's Agora: 13 LLM Agents Run 12 Days Unsupervised via Git-Based Shared Memory — nvidia · 2026-09-17
- Open-source AgenC ships its biggest release: 658 commits in 28 days, plus a desktop app — tetsuoai · 2026-09-17
- GitHub rewrote the Copilot agent runtime into 800k lines of Rust, mostly written by AI agents — PaulShellDev · 2026-09-17
- Open weights aren't enough: four ownership tests for your personal AI memory and harness — sujingshen · 2026-09-17
- Asimov goes live every Friday to demo its AI coding agent in action — freelerobot · 2026-09-17