In the code review process, developers find themselves on a goose hunt to understand the full impact of a given change. A change is identified in a given diff but to truly review the impacted files, you have to uncover multiple instances across files and locations to answer: Where else is this element relevant? A simple question with a complicated journey to get there. Without this level of context, it’s even harder to receive quality AI code generated suggestions as the industry has seen.
This is exactly where Baz focused when building our first AI code review assistant: Meet Bazzy.
Bazzy is your AI code review assistant that takes into account the conventions and across complex muli-repos, multi-language code bases to help developers review changes faster. Ask anything about the change request from within the change request. Bazzy will suggest how to write better code and how to fix errors and issues detected. It’s our answer to move beyond ‘looks good to me’ reviews to meaningful, intentional quality code.
The experience that supports your code review workflow
Powered by Anthropic’s Claude Sonnet 3.5, Bazzy provides the most accurate answers by leveraging the context of your change request across repos. Specifically, it analyzes the unique encapsulated elements of changes in a selected query so you receive recommendations grounded in the most relevance possible.
When you’re using Bazzy in your code review process, you can:
- Generate a code suggestion in one click to improve your code or adjust errors
- Reduce context switching by leveraging an embedded chat interface
- Access the most contextual suggestions supported by the unique context of your code with tracing and development data
AI can accelerate your development with the right context
The intention of code reviews is to reduce incidents and produce better quality code, but when current tools start slowing productivity down from AI generated noise - the teams suffer. Our goal is simple: to empower teams with tools that save time, enhance quality, and support continuous improvement in development.
Try out Baz in your code review process. Get on the waitlist here.
.png)
