Agent-run 'company' needed just 6 compactions in 1.5 months; paper shows benchmarks are compaction-friendly

maksym_andr · x · 2026-11-18

The author notes that Horizon Labs — the AI-run 'company' — operated for 1.5 months on a single Claude Code subscription with only 6 context compactions, plus a daily-updated COMPANYSTATE.md for major events.

The companion paper (arXiv:2610.31778) systematically studies compaction:

Conclusion: most natural tasks are compaction-friendly; compaction + external memory effectively solves continual learning for agents.

Original post →

More from coding & agent

coding & agent channel →