Unblocked Code Review - AI code review that knows when to chime in
by•
AI code review that sees your context, not just your diff. Unblocked draws on context from your whole repo, Slack, Jira, docs, PR history, and more. Every comment moves the conversation forward with cited sources. The result: high-signal comments you'll actually want to implement.
Replies
We've been on the beta for this for several months, and it has been great. False positive rate is near zero, "oh wow, good catch by Unblocked" rate is at least 1 per day. It is much quieter - in the best possible way - than competing tools we've tried, it is almost always right (for us, in our codebase), and it provides actionable feedback. We were delighted when we realized it was sometimes intentionally tagging folks in comments who have worked and commented a lot on code near the change(s) under review, but hadn't otherwise been notified about a PR. This is the most thoughtful, human-friendly UX I've encountered in any AI-enabled tool.
@mfwalters thank you so much Matt, means a ton that you and the team are getting value and took time to post a note here.
We'll keep shipping great product for you and the team 😁
Unblocked
@mfwalters Thanks Matt that means a ton from someone who witnessed our progress in real-time 🥹.
@mfwalters , we appreciate the kind words and all the input and feedback your team has shared along the way!
Congrats on the launch — love the context-first layer powering smarter AI code reviews.
Thanks, @zeiki_yu ! Context makes for great code reviews, just like if a Sr. engineer reviews your diff
Pulling in larger codebase/product/org context makes so much sense. Can't wait to try this out!
@mattfogel thanks! We'd greatly value your feedback on the product considering you've shipped at every scale of company size 😁
@vouchy there's tons of times at previous companies where I've worked on something that shipped only to find out that a late night Slack thread conversation had decided we should no longer do it. Teams are even more dynamic these days with claude code helping to push up PRs so getting the business logic and intent of a code change right is even more important.
@dennispilarinos Congrats on the launch Dennis! How does this handle ambiguous situations where the “right” answer depends on team context or judgment?
Hard part with AI PR review is staying quiet. Only comment when issues are found in Unblocked Code Review is a good default. How do you decide when to chime in, and can teams cap comments per PR? That's the restraint that earns trust.
The "near zero false positive rate" from Matt's review is the real differentiator. Every AI code review tool I've tried drowns you in noise - 50 comments per PR where maybe 2 matter. If you've actually solved signal-to-noise, that's huge.
Curious about the security angle - does it catch things like hardcoded secrets or auth bypasses that slip through when AI-generated PRs come in faster than humans can review them? That's becoming the biggest gap as more teams adopt vibe coding.
Great project! High-quality memory implementation is a common challenge across all AI projects.
Love the “context layer” angle — solves the real pain: agents need shared, team-specific context, not more prompts.