Scaling long-running autonomous coding · Cursor

We've been experimenting with running coding agents autonomously for weeks at a time.

Blog / research We've been experimenting with running coding agents autonomously for weeks. Our goal is to understand how far we can push the frontier of agentic coding for projects that typically take human teams months to complete. This post describes what we've learned from running hundreds of concurrent agents on a single project, coordinating their work, and watching them write over a million lines of code and trillions of tokens. # The limits of a single agent Today's agents work well for focused tasks, but are slow for complex projects. The natural next step is to run multiple agents in

saved by

• Linda
• Aryan Naik
• Serena Ge
• Vincent Cheng
• Vincent Huang
• Devin Plumb
• Lailah Nabegu

related reading

• Towards self-driving codebases · Cursorcursor.com
• Benchmarking Coding Agents on Databricks’ Multi-Million Line Codebasedatabricks.com
• Agent swarms and the new model economics · Cursorcursor.com
• Dynamic context discovery · Cursorcursor.com
• Building Effective AI Agents \ Anthropicanthropic.com
• One Developer, Two Dozen Agents, Zero Alignmentmaggieappleton.com
• Training Composer for longer horizons · Cursorcursor.com
• Effective harnesses for long-running agents \ Anthropicanthropic.com
• Best practices for coding with agents · Cursorcursor.com
• Astra for Coding: Why Are We Doing This Again?lucumr.pocoo.org
• The third era of AI software developmentcursor.com
• Shipping at Inference-Speed | Peter Steinbergersteipete.me