cubic is an AI-powered code review platform that automatically reviews PRs and gives human reviewers superpowers. It’s tool of choice for fast-moving teams like cal.com and n8n.
Subscribe
Sign in
cubic is an AI-powered code review platform that automatically reviews PRs and gives human reviewers superpowers. It’s tool of choice for fast-moving teams like cal.com and n8n.
cubic
Hey Product Hunt, Paul and Allis here – founders of cubic (formerly mrge)! 👋
We’re building cubic – an AI code review platform to help teams ship code faster with fewer bugs. Our early users include Cal.com, n8n, and Better Auth—teams that handle a lot of PRs every day.
🚀 See it in action for cal.com here
We’re engineers who faced this problem when we worked together at our last startup. Code review quickly became our biggest bottleneck and quality tanked — especially as we started using AI to code more.
We had more PRs to review, subtle AI-written bugs slipped, and we (humans) found ourselves rubber-stamping PRs without deeply understanding the changes.
👷 We’re building cubic to help solve that. Here’s how it works:
Connect your GitHub repo via our Github app in two clicks (and optionally download our desktop app).
AI review: When you open a PR, our AI reviews your changes directly in a secure container. It has context into not just that PR, but your whole codebase, so it can pick up patterns and leave comments directly on changed lines. Once the review is done, the sandbox is torn down and your code deleted.
Human-friendly review workflow: Jump into our web (or desktop) app (it’s like Linear but for PRs). Changes are grouped logically (not alphabetically), with important diffs highlighted, visualized, and ready for faster human review.
💻 The AI reviewer works a bit like Cursor in the sense that it navigates your codebase using the same tools a developer would—like jumping to definitions or grepping through code.
⚡️ The platform itself focuses entirely on making *human* code reviews easier. A big inspiration came from productivity-focused apps like Linear or Superhuman, products that show just how much thoughtful design can impact everyday workflows. We wanted to bring that same feeling into code review.
That’s one reason we built a desktop app. It allowed us to deliver a more polished experience, complete with keyboard shortcuts and a snappy interface.
We think the future of coding isn’t about AI replacing humans—it’s about giving us better tools to quickly understand high-level changes, abstracting more and more of the code itself. As code volume continues to increase, this shift is going to become increasingly important.
🚀 cubic is free for 2 weeks. You also get 50% off for 2 months with the code: PHUNT
Just sign up with your Github account to get started!
Looking forward to your feedback—fire away!
@paul_sangle_ferriere1 Amazing work! How does mrge handle very large PRs with cross-file dependencies?
cubic
@masump Great question – under the hood, the AI navigates the codebase just like a real human would (eg grepping files, going to function definitions...) – which is how it handles large PRs. We also run various sub-agents that each specialize in finding specific issues, which helps "break down" large PRs into more manageable checks.
Technically, the way it works is that whenever you open a PR, mrge clones your repository and checks out your branch in a secure and isolated temporary sandbox. We provision this sandbox with shell access and a Language Server Protocol (LSP) server. The AI reviewer then reviews your code, navigating the codebase exactly as a human reviewer would—using shell commands and common editor features like "go to definition" or "find references". When the review finishes, we immediately tear down the sandbox and delete the code—we don’t want to permanently store it for obvious reasons.
@paul_sangle_ferriere1 Kudos! I was wondering if it can be used for reviewing opinionated code style? Maybe use existing cursor/trae rules?
cubic
@rahool_lol Hey Rahul, thanks for the kind words!
Yep, that’s exactly what our custom rules are for. When you set up your account, we suggest a bunch of rules you can use—either based on your existing Cursor/Trae rules or even your team’s past comments in the repo. You can tweak them to match your style, so you’re not stuck with only generic linting.
Congrats on launching mrge, Paul and Allis 🎉 This looks like a smart solution to a growing problem, esp with the rise of Cursor, Windsurf etc. The codebase-aware reviews and logical grouping of changes sound particularly useful. Looking forward to trying this out on our repos!
cubic
Thanks @merlin_k , great to hear that.
Feel free to email / DM if you have any feedback!
SyncSignature
This can be a huge time saver!
cubic
Thanks Neel, great to hear it's resonating!