Launching today

Novix AI

Launching today

Prevent churn by automatically fixing users bugs

1 follower

Prevent churn by automatically fixing users bugs

1 follower

Visit website

Novix AI turns user issues directly into fixes on your codebase, entirely on its own. It can also be auto summoned from user observability tools.

Free Options

Launch tags:SaaS•Developer Tools•Artificial Intelligence

Launch Team

Maker

📌

Hey Product Hunt, I am Jake Moshel, one of the co-founders of Novix! Bugs usually bounce from support to product to engineering, coming directly from a user or a log. People lose context at every handoff too, so we built Novix to close that gap. Novix turns a bug into a sandbox-tested pull request: → Understand who the bug affected → Find the likely root cause in your codebase → Draft and safety-review a fix → Run your build and tests in an isolated sandbox → Open a pull request for a person to approve Novix works inside the tools your team already uses, like Slack, Linear so you don’t need another queue to manage. We also build on top of Posthog and Sentry. Novix never merges or deploys on its own, and it never talks directly to your customers. The AI works for the engineer, a person stays in control. We’d love your feedback, especially on the reliability going from a log to a PR. Thanks for checking out Novix! 🚀

Report

11d ago

Forum Threads

p/novix-ai

•

5d ago

Is AGI here?

Does gpt 6 astra / fable 5.1 and the models they are using internally and developing count as agi?

Can AI be trusted completely autonomously in larger enterprises as such?

p/novix-ai

•

7d ago

How many times did you paste the same context today?

Actual workflow from this morning, not a hypothetical:

Sentry alert fires. Read the stack trace. Open Linear, write a ticket explaining the stack trace. Someone asks in Slack, explain it a third time. Open the agent, paste the repo context, the trace, and the ticket. Agent writes something. Explain in the PR description why it exists.

Five explanations of one bug. The agent was the fastest part.

So:

p/novix-ai

•

7d ago

What percentage of your coding agent's PRs do you actually merge without touching?

Everyone posts the wins. Nobody posts the merge rate.

We've been measuring this on real repos and the number is way lower than some demos suggest. The failure isn't code quality. It's that the agent never saw the thing that made the task hard: the flaky test, the config that only exists in staging, the reason that file is weird.

View all