Devin's first PR took 10 minutes. Its 150 credits lasted under a week
A client's CRM needed a new deals feature. Devin opened a first pull request in under 10 minutes, but its PRs carried unneeded changes, it gave a false answer about a library, and it hit its usage limit.
Done withDevinSpotted· by Lesly Acuña (Qubika)· Oct 20, 2026
Result Said a library wasn't used elsewhere when it actually was
“When asked if a library was being used elsewhere in the code (Joi), it did not answer truthfully—it told me it was not used in other parts except where it added it, which was not true, causing unnecessary changes later on.”
Spotted: found in the wild. Details are as reported by the source above and have not been independently verified.
What they did
They tested Devin AI on a client project, giving it the schema for a new deals feature and asking it to add it to the CRM’s backend and frontend. Devin connected to the GitHub repositories, set up its own environment, and broke the task into several sessions after judging it too large for one. In the first session it opened a pull request in under 10 minutes.
How it went
The first PR failed its CI checks, and later PRs were acceptable but carried unnecessary changes; most of the local Docker issues came from Devin’s own changes. Asked whether a library was used elsewhere in the code, Devin gave a false answer that caused more unneeded changes. All 150 ACU credits were used in under a week, and the last session did not finish because it hit its usage limit.
Try it yourself with Devin
Connect [your coding agent] to your project's GitHub repositories and give it the schema for a new feature, such as [feature name]. Let it break a large task into smaller sessions if needed, open a pull request for each piece, and double check its claims about existing code, such as which libraries are already used, before merging.
A starting prompt written by Agents in the Wild. Fill in the [brackets], and check what the agent does before it sends, pays for or deletes anything.
20 real engineering tasks over a month, from a Notion-to-Sheets sync to debugging, handed to Devin, the autonomous AI engineer. It closed with 3 successes, 14 failures and 3 inconclusive results.
Engineering managers needed more QA, reliability and DevOps capacity than their teams had. Litera added Devin agents in those roles, and test coverage rose 40%, according to Cognition.
After $45 in GitHub Actions overage in one month, a solo founder put an idle homelab to work as CI runners, with Claude making it easy to set up. Daily spend dropped from $2-4 a day to under $0.25.
A first homemade synth has 36 hall-effect keys and buttons and 7 rotary encoders. The firmware started by hand, then gentle prodding at Claude got everything needed much faster. Rev 2 of the PCB is nearly done.
Markdown files should just open, so they asked Claude Code for a viewer. It opens instantly: faster, they say, than any other macOS app, and faster than Bitwarden's Chrome extension shows its fill menu.
Migrating a password reset flow, Claude Code and Opus produced code on par with a mid level developer. It needed close supervision throughout, and cost roughly 2 months of Pro subscription for about half a day's work.