⌘K
Change language Switch ThemeSign In
Narrow Mode
Devin Usage Data: 70% Human-Initiated, Projected to Reach 10/90 Within a Year
Devin Usage Data: 70% Human-Initiated, Projected to Reach 10/90 Within a Year
[](https://www.bestblogs.dev/en/tweets?sourceId=SOURCE_3e922b05)
@hongming731
Scott Wu 分享了 Devin 的使用数据，目前 70%的任务由人类启动，并预测几个月内将颠倒为 30/70，一年内达到 10/90，还描绘了原生智能体开发团队的愿景。
#### Scott Wu
@ScottWu46 · 1d ago
This week 70% of all Devins were started by humans (webapp, slack, linear) and 30% were started automatically (API, and now scheduled + managed Devins)
In a few months that probably flips to 30/70 the other way and within a year it'll be 10/90. What does it look like to run a truly agent-native dev team?
-Handing whole project specs to a top-level agent that manages its subagents to do the groundwork
-Agents that kick off automatically on Sentry/Datadog alerts and as first-line incident-response
-Agents continuously running integration test / QA workflows and auto-investigating UX changes or performance issues
And of course, all of the "eng scaffolding" needed to make this possible - comprehensive unit testing, good documentation, and a reproducible dev environment.Show More
11
19
230
40K
Mar 21, 2026, 11:21 PM View on X
1 Replies
0 Retweets
0 Likes
354 Views 
[](https://www.bestblogs.dev/en/tweets?sourceid=3e922b05) @hongming731
One Sentence Summary
Scott Wu shares Devin usage data showing 70% of tasks are currently human-initiated, projecting a shift to 10/90 automatic initiation within a year, and outlines the vision for an agent-native development team.
Summary
This tweet forwards Scott Wu's (Cognition CEO)分享 about Devin AI coding agent usage data. The core information includes: currently 70% of Devin tasks are initiated by humans through webapp, Slack, and Linear, while 30% are initiated via API, automated scheduling, and management; it's predicted that this will flip to 30/70 within a few months and reach 10/90 within a year. Scott Wu also描绘了原生智能体开发团队的愿景：顶级代理管理子代理完成项目、代理自动响应 Sentry/Datadog 告警、持续运行集成测试和 QA 工作流自动调查 UX 变化或性能问题，同时需要完善的单元测试、良好文档和可复现开发环境等「工程脚手架」。
AI Score
73
Influence Score 1
Published At Yesterday
Language
Chinese
Tags
Devin
AI Programming Agent
Cognition
Scott Wu
Agent HomeArticlesPodcastsVideosTweets