Devin
Autonomous AI software engineering agent from Cognition that writes, tests, and submits code changes.
Liked as a cloud-hosted, team-managed coding agent that works in its own sandboxed environment and opens PRs,12 and several commenters say it has improved a lot.34 The most common complaint is that it struggles with anything beyond trivial work,56 and some find Claude Code or Codex better.78
Why people recommend it
Team-level agent · 2
Seen as a strong cloud-hosted, team-managed agent that works autonomously and can be controlled via API.
“at this point is one of the best cloud-hosted, team-managed coding agents”
“It's a great product compared to Copilot.”
Sandboxed environment · 2
Has its own VM, IDE, terminal and browser to run, test and record changes before sending a PR.
“We use https://devin.ai for this and it works very well.”
“at this point is one of the best cloud-hosted, team-managed coding agents”
Much improved · 3
Former skeptics and long-time users say it has matured and is now fairly good.
“I used to be a hater, but it’s kinda good now”
“at this point is one of the best cloud-hosted, team-managed coding agents”
Show 1 more
“I use it, and have been happy with it for the most part.”
Beats Copilot · 1
An enterprise user calls it a great product compared to Copilot.
“It's a great product compared to Copilot.”
Caveats and dissent
Weak results · 2
Some found it poor even after re-evaluation, or only useful for trivial tasks.
“It sucked the second time also...”
“most people can't even get the tools to do anything useful more than trivial things”
Behind competitors · 2
Some users find Claude Code and Codex better, and question its hands-off PR approach.
“It's a great product compared to Copilot.”
“I cant believe people still use Devin.”
Inconsistent clients · 1
Web and desktop clients use different harnesses, so response quality varies greatly.
“It's a great product compared to Copilot.”
Black box · 1
Criticized as too opaque and prone to hallucinating on complex legacy code.
“I found existing AI coding tools (like Claude Code or Devin) too "black box."”
Vendor lock-in · 1
One commenter warns against depending on a single provider like Devin.
“you should be free and not tied to a single provider (e.g. Devin)”
Often compared with
Who recommends it in their own words
Threads where it came up
- Cognition launches new SWE-2 model, Rivaling Fable 5.1 and GPT-Astra 5 mentions
- Open Oncall: Open-Source Alternative to Devin Oncall 1 mention
- RSA-896 1 mention
- GPT-5.6 Luna vs. GPT-6 Astra: Is a $1.20 Model Good Enough for Code Review? 1 mention
- Cognition (Devin) raises $2B at $48B valuation 1 mention
- SpaceX to buy Cursor AI coding agent operator Anysphere for $60B 1 mention
Data: devin.json. Quotes are verbatim; the verdict and theme labels are written by a model from these comments, and every clause links to its sources.
All comments 14
“you should be free and not tied to a single provider (e.g. Devin)”
“Devin (their AI agent) ported CADO-NFS to run on GPUs”
“finds bugs missed by coderabitai, devin-integration-bot, codex and copilot”
“at this point is one of the best cloud-hosted, team-managed coding agents”
“It sucked the second time also...”
“It's a great product compared to Copilot.”
“I use it, and have been happy with it for the most part.”
“I pay for it (mostly because they grandfathered me from the old prices)!”
“I used to be a hater, but it’s kinda good now”
“I cant believe people still use Devin.”
“I found existing AI coding tools (like Claude Code or Devin) too "black box."”
“We use https://devin.ai for this and it works very well.”
“drafting a prompt for an AI agent (Cursor/Devin/Codex/etc) and letting it run”
“most people can't even get the tools to do anything useful more than trivial things”