Devin: what 16 developers actually say.
16 Hacker News opinions from 16 developers, Nov 2025 – Nov 2026. 9 come from people who say they use it. Here is what they praise, what they complain about, and who switched.
The case for Devin, in one paragraph.
Devin earns 28% approval across 16 opinions. Developers are divided on it.
Pick it if you
- Want to run many cloud agents in parallel across a team
- Already use ACP-compatible harnesses and want them in one desktop app
- Value the UX of team-managed agents and can absorb the cost
Look elsewhere if you
- Need onboarding that gets an existing project running without handholding
- Are cost-sensitive or plan to run many automations
- Expect reliable output on long refactors with little supervision
Where it wins and loses.
Mentions across 16 opinions. Bars are counts, not votes; a tool can be both praised and criticized on the same topic.
View data table
| Topic | Praise | Complaints | Mentions |
|---|
How opinion has moved.
Monthly approval, Feb 2026 – Sep 2026. Bars show opinion volume; low-volume months are still included but faded.
Not enough monthly data for a trend yet.
What developers say.
Summaries are paraphrased. Each links to the original Hacker News comment.
Praise
2 selectedDevin Desktop supports most ACP-compatible harnesses, so it can run Claude Code, Codex, or OpenCode.
Cloud agents like Devin let teams spawn many agents at once, giving teams that use them an edge.
Complaints
4 selectedCalls Devin CLI a horrible harness that lacks a scheduling system.
Devin failed to get their project running after an hour of setup, pointing to weak onboarding.
Spent two days on a transpiler with Devin; it got confused, duplicated code, and was abandoned.
Calls Devin overpriced and ahead of its time for enterprise use.
Questions developers ask.
Is Devin worth it?
Is Devin better than OpenAI Codex?
Why do developers complain about Devin?
The data behind this page.
Developers give Devin 28% approval across 16 opinions from 16 authors, and 9 of those authors used it firsthand. Complaints center on cost, weak onboarding, and code quality: one user abandoned a two-day transpiler task after duplicated code, and others call it overpriced for enterprise use. The autonomy aspect draws 1 positive and 2 negative mentions. The recent trend is modestly better: September 2026 accounts for 6 of the 16 opinions at 42% approval, up from 0% in several earlier months, though monthly samples are small.
Every number comes from public Hacker News comments collected through the HN Search API. A language model tagged each comment for sentiment, first-hand use, topic and switches; a human spot-checks samples. No vendor can pay for placement.
Get the monthly report
Which AI coding tools gained or lost developer approval on Hacker News, and why people switched. One email a month.