Building recursive agent systems
x.com · 394 words · saved by 1 readers
https://t.co/6nWAu8JVdw
At Cursor, we run thousands of agents to help us train the next version of Composer. We give them research tasks, and if they aren't succeeding or run into issues, they DM us on Slack or page us via PagerDuty. Scaling training for Composer We’ve built an org chart of agents that work together. As we’ve scaled training for Composer, we’ve wanted to run thousands more experiments. This was possible before, but it was slow and hard to keep track of every experiment’s status. To speed things up and parallelize work, we built an always-running agent system (yes, it's a loop). An agent system…
saved by
related reading
- Composer2.pdfcursor.com
- How we built our multi-agent research system \ Anthropicanthropic.com
- Automated Weak-to-Strong Researcheralignment.anthropic.com
- Scaling long-running autonomous coding · Cursorcursor.com
- Towards self-driving codebases · Cursorcursor.com
- Trending Papers - Hugging Facepaperswithcode.com
- GitHub - karpathy/autoresearch: AI agents running research on single-GPU nanochat training automaticallygithub.com
- Kimi Agent Swarm: 100 Sub-Agents at Scalekimi.com
- Agent swarms and the new model economics · Cursorcursor.com
- AddyOsmani.com - Long-running Agentsaddyosmani.com
- Building Effective AI Agents \ Anthropicanthropic.com
- Sonya Huang 🐥 on X: "Today's Training Data episode takes us BTS on the infrastructure challenges required to do large RL runs at scale, featuring @ellev3n11 (Composer Lead at @cursor_ai) and @dzhulgakov (Co-Founder at @FireworksAI_HQ). The Cursor team trained Composer 2 on Fireworks by starting with https://t.co/6LLlJlyl8Q" / Xx.com