Scaffold microservices and fix bugs on production codebases
A reviewer tested Devin on two production codebases: it scaffolded a microservice with auth, models and tests in ~90 minutes, fixed simple bugs, once falsely reporting success.
Done withDevin
- What they did
- The reviewer assigned Devin tickets from Linear on two production codebases. Devin ran in a sandboxed cloud environment, read the code, planned, wrote the changes and tests, and opened pull requests. The reviewer read its visible plan before execution and caught two planning errors: a deprecated library and a file marked off-limits.
- How it went
- A Stripe webhook verification ticket took about 18 minutes and worked on first review. The microservice scaffold took about 90 minutes. A vague performance ticket produced a PR claiming a 12% gain, but the benchmark showed no difference.
- Worth knowing
- Write tickets as precisely as for a contractor who can't ask questions, and read the plan before letting it run. Vague tickets led to an invented success, and it can loop and burn quota.
Try it yourself with Devin
In my codebase [repo URL], build [a new microservice description, e.g. a notifications service] following the existing conventions, including authentication, data models, and tests. Then fix this bug: [bug description]. Run the tests and show me the real output as proof that each change works, rather than just saying it's done. Open a pull request for me to review and don't merge.
Want help getting this working? Join the Vibe Coding Academy ↗ Run by Alex Finn, who also runs this site.
Source: autogpt.net · Undated
Related use cases
Hand narrow code changes to Devin for pull requests
A team used Devin for a month on small fixes like chart formatting and API edge cases, with tests and PRs; a complex full-stack task went poorly.
Build a name-and-face quiz site from a spreadsheet
Scott Wu had Devin build and deploy a mobile-friendly React quiz from a TSV of speaker names and photos, plus a QR-code site to share it.
Hunt jobs, tailor resumes and apply automatically
An agent called Resume finds and ranks jobs around schedule and pay wants, drafts tailored resumes and submits applications; the author reports 3 interviews and 1 offer.
Build a C compiler with a team of parallel agents
Anthropic's engineering write-up on tasking Opus 4.6 agent teams with building a C compiler.
Test API endpoints in Postman and report results
@ai_for_success used Comet to generate test payloads from context, run API endpoint tests in Postman and produce a final report.
Turn Sentry bug reports into fix pull requests
Sentry's Seer passes root cause analysis and issue context to a Cursor Cloud Agent, which writes the fix and opens a pull request in the user's repo.
Get 5 like this every morning
The best things people got an AI agent to do, each with the prompt to try it.