Anthropic engineer recounts scaling test impact analysis under agentic coding pressure
Original title:Agentic coding is straining CI. Here’s how we scaled test impact analysis at Anthropic
AI summary
An Anthropic engineer wrote about scaling the test impact analysis service under agentic coding pressure. Claude writes about 80% of the code, the number of tests grew 10x, and CI jobs increased 25x in six months. The post reviews three temporary patches—scaling up, sharding by package, and daily restarts—that lasted only 70 days, 29 days, and less than a day, respectively, before a three-week redesign into a stateless, horizontally scalable architecture with in-memory storage and a journal; it recommends teams plan capacity for 25x load within two quarters.
Key Takeaways
- An Anthropic engineer wrote about scaling the test impact analysis service under agentic coding…
- Claude writes about 80% of the code,the number of tests grew 10x,and CI jobs increased 25x in six…
- The post reviews three temporary patches—scaling up,sharding by package,and daily restarts—that…
- it recommends teams plan capacity for 25x load within two quarters.
Quick Facts
- Published
- Updated
- Source
- Claude:Blog(网页)
- Category
- Opinion
- AI score
- 66
作者复盘了智能体编程把 CI 推向 25x 负载的完整演进,给出可迁移的架构与容量规划经验。
Related entities
Source & copyright
This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.
How to Cite
When citing this article, keep the title, date, and permalink, and credit Orange News as the source.
Orange News: Anthropic engineer recounts scaling test impact analysis under agentic coding pressure. September 14, 2026. https://new.nat6.net/en/news/cmu1miqz1154mrocnhtw6icqw/