Warp 分享软件工厂实践：每个 PR 成本从约 $80 降到 $15-20
⚡ Inside the 2026 SOTA of Software Factories with @zachlloydtweets of @warpdotdev! Warp's open sour...
Warp 的 Zach Lloyd 讲怎么把 agent 写代码的成本从 80 美元压到 15-20 美元一个 PR，还开源了整套流程，做工程的人可以照抄。
Warp 的 Zach Lloyd 在 Latent Space 播客中介绍了其开源的软件工厂方案。通过调优模型和减少冗余 subagent，单 PR 成本从约 $80 降到 $15-20，吞吐量较上月提升 40%。他演示了从 Slack 请求自动生成 PR 的完整流程，并讲了如何度量成本、吞吐量和质量三个指标。Warp 团队还让 agent 来改进工厂本身，并对自有任务做基准测试。
⚡ Inside the 2026 SOTA of Software Factories with @zachlloydtweets of @warpdotdev! Warp's open sour...
⚡ Inside the 2026 SOTA of Software Factories with @zachlloydtweets of @warpdotdev ! Warp's open source cloud factory approach has been resonating bigly this year, leading to multiple top @aiDotEngineer talks and an invitation to keynote this year's AIE CODE. Zach talks through the philosophy, demos Warp’s live system, and reports that tuning models and reducing redundant subagents can lower costs from ~$80 to ~$15–20 per PR and raise throughput 40% over the prior month, while keeping quality! Full episode below: from a Slack request to a PR, then into the work of measuring and improving software factories. 0:00 Software factories at Warp 0:53 AIE CODE in San Francisco 1:31 Zach Lloyd and software factories 4:30 Define the factory in code 7:34 Follow a Slack request to a PR 8:41 Measure cost, throughput and quality 14:32 Agents improving the factory 16:10 Benchmark your own tasks 23:59 From $80 to $15–20 per PR 27:03 Who needs software factories? 33:23 Product engineers and factory engineers See Zach at @aiDotEngineer CODE, Nov 10–12 in San Francisco: ai.engineer/code Your browser does not support the video tag. 🔗 View on Twitter 💬 1 🔄 1 ❤️ 1 👀 482 📊 2 ⚡
- Replit11-03 18:44原文
