{"type":"analysis","id":"analysis:research-coding-agents-finish-long-tasks-when-someone-else-decides-when-they-re-done","slug":"research-coding-agents-finish-long-tasks-when-someone-else-decides-when-they-re-done","title":"Coding agents finish long tasks when someone else decides when they're done","description":"Factory reports that splitting coding agents into an implementer and an independent validator lifted rebuilds of hard ProgramBench programs. GDAL went from 36% to 90% test parity, at about 13 times the wall-clock time. The \"Factory Benchmarks\" item in this cluster is a different product from a different company: Warp's custom eval tool.","published_at":"2026-10-23T17:00:00.000Z","updated_at":"2026-10-23T17:00:00.000Z","record_date":"2026-10-23","topics":["agents","benchmarks","coding_agents","research"],"source_urls":["https://factory.com/news/what-it-takes-for-coding-agents-to-complete-large-software-tasks","https://arxiv.org/abs/2605.03546","https://www.vals.ai/benchmarks/programbench","https://arxiv.org/abs/2607.27167","https://www.warp.dev/blog/warp-factory-benchmarks","https://www.warp.dev/factories/benchmarks"],"language":"en","routes":{"html":"https://newruntime.com/analysis/research-coding-agents-finish-long-tasks-when-someone-else-decides-when-they-re-done/","markdown":"https://newruntime.com/analysis/research-coding-agents-finish-long-tasks-when-someone-else-decides-when-they-re-done.md","json":"https://newruntime.com/analysis/research-coding-agents-finish-long-tasks-when-someone-else-decides-when-they-re-done.json"},"author":{"role":"author","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]},"editorial_provenance":{"desk":"authorial","target_ref":"deep_research:86f89329-c7d0-4277-9f1b-83b7268c8f2d","authorial_article_id":"86f89329-c7d0-4277-9f1b-83b7268c8f2d"},"next_reads":[{"type":"topic","path":"/topics/agents/","url":"https://newruntime.com/topics/agents/","title":"Agents","reason":"Explore the agents topic hub.","media_type":"text/html","markdown_url":"https://newruntime.com/topics/agents.md","json_url":"https://newruntime.com/topics/agents.json"},{"type":"topic","path":"/topics/benchmarks/","url":"https://newruntime.com/topics/benchmarks/","title":"Benchmarks","reason":"Explore the benchmarks topic hub.","media_type":"text/html","markdown_url":"https://newruntime.com/topics/benchmarks.md","json_url":"https://newruntime.com/topics/benchmarks.json"},{"type":"topic","path":"/topics/coding-agents/","url":"https://newruntime.com/topics/coding-agents/","title":"Coding Agents","reason":"Explore the coding_agents topic hub.","media_type":"text/html","markdown_url":"https://newruntime.com/topics/coding-agents.md","json_url":"https://newruntime.com/topics/coding-agents.json"},{"type":"collection","path":"/analysis/","url":"https://newruntime.com/analysis/","title":"Analysis","reason":"Open the latest authorial analysis.","media_type":"text/html","json_url":"https://newruntime.com/analysis.json"},{"type":"collection","path":"/source-ledger.json","url":"https://newruntime.com/source-ledger.json","title":"Source Ledger","reason":"Inspect the public source evidence graph.","media_type":"application/json"}]}