News for hackers. Stuff that matters.

vishuk

Kudos to CodeRabbit!

Squeezing Blood from a Stone: Overcoming LLM Limitations

https://coderabbit.ai/blog/coderabbit-deep-dive

Hello readers! Our team has built an AI/LLM-driven code review tool that significantly helps improve dev velocity and code quality.We are sharing a blog post covering our journey to build CodeRabbit, how we worked around AI/LLM limitations, and our vision to see CodeRabbit as an AI-first developer tools company.Please look at it. We will happily answer any technical questions regarding the...

( Read More... | no comments | 2 points | on HN )

Show HN: AI and the Future of Code Reviews

https://coderabbit.ai/blog/coderabbit-deep-dive

Hello readers! Our team has built an AI-driven code review tool for GitHub PRs leveraging OpenAI’s gpt-3.5-turbo and gpt-4 models. The tool significantly helps improve dev velocity and code quality.We also wrote a blog post covering our journey to build CodeRabbit, how we worked around AI’s limitations, and our vision to see CodeRabbit as an AI-first developer tools company.Please take a look...

( Read More... | no comments | on HN )
Re:AI/LLM driven GitHub PR reviewer on Monday August 07, 2023 @03:40AM (#37031466)
Hello HN readers! Our team has built an AI/LLM-driven code review tool that significantly helps improve dev velocity and code quality. Its unique features are: - Line-by-line code change suggestions: Reviews the changes line by line and provides code change suggestions that can be directly committed from the GitHub UI. - Continuous, incremental reviews: Reviews are performed on each commit within a pull request rather than a one-time review on the entire pull request. - Cost-effective and reduced noise: Incremental reviews reduce noise by tracking changed files between commits and the base of the pull request. - Chat with the bot: Supports conversation with the bot in the context of lines of code or entire files, helpful in providing context, generating test cases, and reducing code complexity. - Smart review skipping: By default, skips in-depth review for simple changes (e.g., typo fixes) and when changes look good for the most part.

We would love the HN community to try it out in their GitHub repos and provide feedback! We will happily answer any technical questions regarding the sophisticated prompt engineering we did for this project.

Re:Amazon Managed Service for Prometheus on Tuesday December 15, 2020 @03:48PM (#25435054)
Do we know which scalable prometheus backend are they running? Chronosphere? Thanos?