CodeRabbit
@coderabbitai.bsky.social
Supercharge your dev team with AI code reviews. Get your free trial now: https://coderabbit.ai!
Pinned
AI coding assistants are shipping more code...but also more bugs!
Our State of AI vs. Human Code Generation Report analyzed 470 real GitHub PRs.
The findings? AI‑coauthored PRs produced about 1.7× more issues than human‑only PRs!
Our State of AI vs. Human Code Generation Report analyzed 470 real GitHub PRs.
The findings? AI‑coauthored PRs produced about 1.7× more issues than human‑only PRs!
North Pole postmortem just released.
Root cause: elves refactored the gift sorter “to make it more responsive,” introduced a textbook SQL injection, and one kid briefly qualified for 48,000+ presents.
Mitigation: mandatory CodeRabbit on every PR.
Full postmortem below.
Root cause: elves refactored the gift sorter “to make it more responsive,” introduced a textbook SQL injection, and one kid briefly qualified for 48,000+ presents.
Mitigation: mandatory CodeRabbit on every PR.
Full postmortem below.
December 22, 2025 at 8:15 PM
North Pole postmortem just released.
Root cause: elves refactored the gift sorter “to make it more responsive,” introduced a textbook SQL injection, and one kid briefly qualified for 48,000+ presents.
Mitigation: mandatory CodeRabbit on every PR.
Full postmortem below.
Root cause: elves refactored the gift sorter “to make it more responsive,” introduced a textbook SQL injection, and one kid briefly qualified for 48,000+ presents.
Mitigation: mandatory CodeRabbit on every PR.
Full postmortem below.
Announcing "Share the Bug" CodeRabbit Holiday giveaway! 🗣️
To celebrate the end of a great year, we are giving away:
> 1-year of individual CodeRabbit Pro
> 1-year subscription of Frontend Masters
> 1-year subscription to Egghead.io
To 10 people!
Entry rules in the thread! 👇
To celebrate the end of a great year, we are giving away:
> 1-year of individual CodeRabbit Pro
> 1-year subscription of Frontend Masters
> 1-year subscription to Egghead.io
To 10 people!
Entry rules in the thread! 👇
December 22, 2025 at 4:44 PM
Announcing "Share the Bug" CodeRabbit Holiday giveaway! 🗣️
To celebrate the end of a great year, we are giving away:
> 1-year of individual CodeRabbit Pro
> 1-year subscription of Frontend Masters
> 1-year subscription to Egghead.io
To 10 people!
Entry rules in the thread! 👇
To celebrate the end of a great year, we are giving away:
> 1-year of individual CodeRabbit Pro
> 1-year subscription of Frontend Masters
> 1-year subscription to Egghead.io
To 10 people!
Entry rules in the thread! 👇
Most AI dev tools obsess over vanity metrics like lines of code and accepted completions.
The only metrics that matter:
> What actually ships
> Does it break production.
Our new analytics dashboard puts this visibility directly into the hands of engineering leaders!
The only metrics that matter:
> What actually ships
> Does it break production.
Our new analytics dashboard puts this visibility directly into the hands of engineering leaders!
December 19, 2025 at 5:04 PM
Most AI dev tools obsess over vanity metrics like lines of code and accepted completions.
The only metrics that matter:
> What actually ships
> Does it break production.
Our new analytics dashboard puts this visibility directly into the hands of engineering leaders!
The only metrics that matter:
> What actually ships
> Does it break production.
Our new analytics dashboard puts this visibility directly into the hands of engineering leaders!
2025 was the year the internet kept breaking 🔥 🌐
Outages are up, incidents are up, and AI-generated code is a big part of the story!
AI code now ships with around 1.7x more issues and more high‑risk vulnerabilities than human-written code.
Outages are up, incidents are up, and AI-generated code is a big part of the story!
AI code now ships with around 1.7x more issues and more high‑risk vulnerabilities than human-written code.
December 18, 2025 at 4:33 PM
2025 was the year the internet kept breaking 🔥 🌐
Outages are up, incidents are up, and AI-generated code is a big part of the story!
AI code now ships with around 1.7x more issues and more high‑risk vulnerabilities than human-written code.
Outages are up, incidents are up, and AI-generated code is a big part of the story!
AI code now ships with around 1.7x more issues and more high‑risk vulnerabilities than human-written code.
Reposted by CodeRabbit
Code review has quietly become a new bottleneck.
AI helps devs ship more code which means more PRs, more risk, and less time to review carefully.
So how about some AI Code Review to help with all the AI generated code? That's exactly what @coderabbitai.bsky.social built.
expo.dev/blog/ai-powe...
AI helps devs ship more code which means more PRs, more risk, and less time to review carefully.
So how about some AI Code Review to help with all the AI generated code? That's exactly what @coderabbitai.bsky.social built.
expo.dev/blog/ai-powe...
AI-powered code reviews for your Expo projects
CodeRabbit is an AI-powered code review platform that understands Expo. It catches real issues across your whole codebase before they hit production.
expo.dev
December 17, 2025 at 4:36 PM
Code review has quietly become a new bottleneck.
AI helps devs ship more code which means more PRs, more risk, and less time to review carefully.
So how about some AI Code Review to help with all the AI generated code? That's exactly what @coderabbitai.bsky.social built.
expo.dev/blog/ai-powe...
AI helps devs ship more code which means more PRs, more risk, and less time to review carefully.
So how about some AI Code Review to help with all the AI generated code? That's exactly what @coderabbitai.bsky.social built.
expo.dev/blog/ai-powe...
AI coding assistants are shipping more code...but also more bugs!
Our State of AI vs. Human Code Generation Report analyzed 470 real GitHub PRs.
The findings? AI‑coauthored PRs produced about 1.7× more issues than human‑only PRs!
Our State of AI vs. Human Code Generation Report analyzed 470 real GitHub PRs.
The findings? AI‑coauthored PRs produced about 1.7× more issues than human‑only PRs!
December 17, 2025 at 4:49 PM
AI coding assistants are shipping more code...but also more bugs!
Our State of AI vs. Human Code Generation Report analyzed 470 real GitHub PRs.
The findings? AI‑coauthored PRs produced about 1.7× more issues than human‑only PRs!
Our State of AI vs. Human Code Generation Report analyzed 470 real GitHub PRs.
The findings? AI‑coauthored PRs produced about 1.7× more issues than human‑only PRs!
The CodeRabbit BLR team out in full force for the Microsoft AI Tour last Friday!
A whole day of helping people learn to transform their orgs with the power of AI!
🇮🇳🐰
A whole day of helping people learn to transform their orgs with the power of AI!
🇮🇳🐰
December 15, 2025 at 5:16 PM
The CodeRabbit BLR team out in full force for the Microsoft AI Tour last Friday!
A whole day of helping people learn to transform their orgs with the power of AI!
🇮🇳🐰
A whole day of helping people learn to transform their orgs with the power of AI!
🇮🇳🐰
If anyone knows how challenging manual code reviews are, it’s The Linux Foundation.
Before CodeRabbit, they dealt with:
+ Slow merge time
+ Hidden bugs
+ Low dev productivity
Now, they’re sharing exactly why they love CodeRabbit in our case study. 👇
Before CodeRabbit, they dealt with:
+ Slow merge time
+ Hidden bugs
+ Low dev productivity
Now, they’re sharing exactly why they love CodeRabbit in our case study. 👇
December 10, 2025 at 4:11 PM
If anyone knows how challenging manual code reviews are, it’s The Linux Foundation.
Before CodeRabbit, they dealt with:
+ Slow merge time
+ Hidden bugs
+ Low dev productivity
Now, they’re sharing exactly why they love CodeRabbit in our case study. 👇
Before CodeRabbit, they dealt with:
+ Slow merge time
+ Hidden bugs
+ Low dev productivity
Now, they’re sharing exactly why they love CodeRabbit in our case study. 👇
"Debugging is twice as hard as writing the code in the first place. Therefore, if you write the code as cleverly as possible, you are, by definition, not smart enough to debug it."
- Brian Kernighan (co-creator of Unix)
Is it even hard with AI involved?
Let's talk about it 👇
- Brian Kernighan (co-creator of Unix)
Is it even hard with AI involved?
Let's talk about it 👇
December 4, 2025 at 4:47 PM
"Debugging is twice as hard as writing the code in the first place. Therefore, if you write the code as cleverly as possible, you are, by definition, not smart enough to debug it."
- Brian Kernighan (co-creator of Unix)
Is it even hard with AI involved?
Let's talk about it 👇
- Brian Kernighan (co-creator of Unix)
Is it even hard with AI involved?
Let's talk about it 👇
The Vibe Code Cleanup Specialists have landed at AWS RE:Invent 2025!
If you're suffering from dirty code, drop by the booth, chat to the team, and grab some clean merch 🤙🫧
If you're suffering from dirty code, drop by the booth, chat to the team, and grab some clean merch 🤙🫧
December 1, 2025 at 9:32 PM
The Vibe Code Cleanup Specialists have landed at AWS RE:Invent 2025!
If you're suffering from dirty code, drop by the booth, chat to the team, and grab some clean merch 🤙🫧
If you're suffering from dirty code, drop by the booth, chat to the team, and grab some clean merch 🤙🫧
Open source is booming, but so is maintainer burnout.
Langflow turned to CodeRabbit to keep PR quality high, cut review bottlenecks, and make shipping faster (and saner) for their founder led team.
Find out why Langflow loves CodeRabbit in our latest case study! 👇
Langflow turned to CodeRabbit to keep PR quality high, cut review bottlenecks, and make shipping faster (and saner) for their founder led team.
Find out why Langflow loves CodeRabbit in our latest case study! 👇
December 1, 2025 at 7:31 PM
Open source is booming, but so is maintainer burnout.
Langflow turned to CodeRabbit to keep PR quality high, cut review bottlenecks, and make shipping faster (and saner) for their founder led team.
Find out why Langflow loves CodeRabbit in our latest case study! 👇
Langflow turned to CodeRabbit to keep PR quality high, cut review bottlenecks, and make shipping faster (and saner) for their founder led team.
Find out why Langflow loves CodeRabbit in our latest case study! 👇
If GPT-5.1 is the decisive teammate and Opus 4.5 the disciplined architect, Gemini 3 is the dense engineer who delivers a fully reasoned diff that is confident, comprehensive, and intent on proving its point.
November 26, 2025 at 7:37 PM
If GPT-5.1 is the decisive teammate and Opus 4.5 the disciplined architect, Gemini 3 is the dense engineer who delivers a fully reasoned diff that is confident, comprehensive, and intent on proving its point.
The bottleneck in software development has shifted from writing code to validating it 🧪
84% of devs trust AI tools, but 48% still distrust the accuracy!
Our latest blog post goes into all things AI code trust, and how we aim to solve it!
84% of devs trust AI tools, but 48% still distrust the accuracy!
Our latest blog post goes into all things AI code trust, and how we aim to solve it!
November 25, 2025 at 6:51 PM
The bottleneck in software development has shifted from writing code to validating it 🧪
84% of devs trust AI tools, but 48% still distrust the accuracy!
Our latest blog post goes into all things AI code trust, and how we aim to solve it!
84% of devs trust AI tools, but 48% still distrust the accuracy!
Our latest blog post goes into all things AI code trust, and how we aim to solve it!
We benchmarked Opus 4.5 on coding tasks!
We found it equals GPT 5.1 in many respects and is an improvement over Sonnet 4.5.
If Sonnet 4.5 feels like a teacher and GPT-5.1 like a teammate, Opus 4.5 is the architect reviewing your PR!
We found it equals GPT 5.1 in many respects and is an improvement over Sonnet 4.5.
If Sonnet 4.5 feels like a teacher and GPT-5.1 like a teammate, Opus 4.5 is the architect reviewing your PR!
November 24, 2025 at 7:05 PM
We benchmarked Opus 4.5 on coding tasks!
We found it equals GPT 5.1 in many respects and is an improvement over Sonnet 4.5.
If Sonnet 4.5 feels like a teacher and GPT-5.1 like a teammate, Opus 4.5 is the architect reviewing your PR!
We found it equals GPT 5.1 in many respects and is an improvement over Sonnet 4.5.
If Sonnet 4.5 feels like a teacher and GPT-5.1 like a teammate, Opus 4.5 is the architect reviewing your PR!
CodeRabbit lets you connect multiple MCP servers for smarter code reviews!
But why should you?
> Enrich code reviews with context from multiple tools.
> Make informed and precise reviews.
Check out our latest blog to follow along and learn how to set up a Slack MCP server!
But why should you?
> Enrich code reviews with context from multiple tools.
> Make informed and precise reviews.
Check out our latest blog to follow along and learn how to set up a Slack MCP server!
November 21, 2025 at 4:43 PM
CodeRabbit lets you connect multiple MCP servers for smarter code reviews!
But why should you?
> Enrich code reviews with context from multiple tools.
> Make informed and precise reviews.
Check out our latest blog to follow along and learn how to set up a Slack MCP server!
But why should you?
> Enrich code reviews with context from multiple tools.
> Make informed and precise reviews.
Check out our latest blog to follow along and learn how to set up a Slack MCP server!
Introducing Agentic PR Summary!
Need more from your CodeRabbit Summary?
Now you can customize it with the new High Level Summary Instructions setting ✌️
> Auto fill your Pull Request Template.
> Format it how you want: bullets, tables, sections.
+ Add custom instructions!
Need more from your CodeRabbit Summary?
Now you can customize it with the new High Level Summary Instructions setting ✌️
> Auto fill your Pull Request Template.
> Format it how you want: bullets, tables, sections.
+ Add custom instructions!
November 19, 2025 at 4:29 PM
Introducing Agentic PR Summary!
Need more from your CodeRabbit Summary?
Now you can customize it with the new High Level Summary Instructions setting ✌️
> Auto fill your Pull Request Template.
> Format it how you want: bullets, tables, sections.
+ Add custom instructions!
Need more from your CodeRabbit Summary?
Now you can customize it with the new High Level Summary Instructions setting ✌️
> Auto fill your Pull Request Template.
> Format it how you want: bullets, tables, sections.
+ Add custom instructions!
Shipping auth and user-management infra means zero tolerance for security lapses or spec violations.
After adding CodeRabbit, Clerk managed to:
+ Cut time to merge by 40%
+ Catch 70% of edge-case bugs before human reviews
Check out the full case study to see how they did it!
After adding CodeRabbit, Clerk managed to:
+ Cut time to merge by 40%
+ Catch 70% of edge-case bugs before human reviews
Check out the full case study to see how they did it!
November 18, 2025 at 4:24 PM
Shipping auth and user-management infra means zero tolerance for security lapses or spec violations.
After adding CodeRabbit, Clerk managed to:
+ Cut time to merge by 40%
+ Catch 70% of edge-case bugs before human reviews
Check out the full case study to see how they did it!
After adding CodeRabbit, Clerk managed to:
+ Cut time to merge by 40%
+ Catch 70% of edge-case bugs before human reviews
Check out the full case study to see how they did it!
We are open-sourcing one of our internal tools!
Introducing our AI-native open-source Git worktree manager CLI that works with all your AI Coding Agents ✌️ (that was a mouth full)
Have fun!
Introducing our AI-native open-source Git worktree manager CLI that works with all your AI Coding Agents ✌️ (that was a mouth full)
Have fun!
November 16, 2025 at 5:23 PM
We are open-sourcing one of our internal tools!
Introducing our AI-native open-source Git worktree manager CLI that works with all your AI Coding Agents ✌️ (that was a mouth full)
Have fun!
Introducing our AI-native open-source Git worktree manager CLI that works with all your AI Coding Agents ✌️ (that was a mouth full)
Have fun!
The CodeRabbit team are getting ready for Microsoft Ignite next week!
Catch us at booth 5314 🧽🧹
Catch us at booth 5314 🧽🧹
November 16, 2025 at 3:16 AM
The CodeRabbit team are getting ready for Microsoft Ignite next week!
Catch us at booth 5314 🧽🧹
Catch us at booth 5314 🧽🧹
Looking for the best Microsoft Ignite after-party?
We're hosting a social rooftop mixer with Anthropic on Wednesday, Nov 19th for AI engineers and leaders!
Swing by the CodeRabbit HQ to enjoy a relaxed evening and raise to toast to the 1 year anniversary of MCP!
🥂
We're hosting a social rooftop mixer with Anthropic on Wednesday, Nov 19th for AI engineers and leaders!
Swing by the CodeRabbit HQ to enjoy a relaxed evening and raise to toast to the 1 year anniversary of MCP!
🥂
November 15, 2025 at 2:19 AM
Looking for the best Microsoft Ignite after-party?
We're hosting a social rooftop mixer with Anthropic on Wednesday, Nov 19th for AI engineers and leaders!
Swing by the CodeRabbit HQ to enjoy a relaxed evening and raise to toast to the 1 year anniversary of MCP!
🥂
We're hosting a social rooftop mixer with Anthropic on Wednesday, Nov 19th for AI engineers and leaders!
Swing by the CodeRabbit HQ to enjoy a relaxed evening and raise to toast to the 1 year anniversary of MCP!
🥂
GPT-5.1 isn’t “GPT-5 but faster.”
In our evals of the model, we found it’s the highest-precision model we’ve ever tested for code-related tasks like code review.
Less noise, more fixes, reviews that read like patches again.
In our evals of the model, we found it’s the highest-precision model we’ve ever tested for code-related tasks like code review.
Less noise, more fixes, reviews that read like patches again.
November 13, 2025 at 6:21 PM
GPT-5.1 isn’t “GPT-5 but faster.”
In our evals of the model, we found it’s the highest-precision model we’ve ever tested for code-related tasks like code review.
Less noise, more fixes, reviews that read like patches again.
In our evals of the model, we found it’s the highest-precision model we’ve ever tested for code-related tasks like code review.
Less noise, more fixes, reviews that read like patches again.
Our very own Director of AI, David Locker, will be taking to the stage at AI Dev 25 x NYC this Friday!
He'll be talking all things CodeRabbit, Context Engineering & Prompt Engineering!
🐰🎙️
He'll be talking all things CodeRabbit, Context Engineering & Prompt Engineering!
🐰🎙️
November 12, 2025 at 4:12 PM
Our very own Director of AI, David Locker, will be taking to the stage at AI Dev 25 x NYC this Friday!
He'll be talking all things CodeRabbit, Context Engineering & Prompt Engineering!
🐰🎙️
He'll be talking all things CodeRabbit, Context Engineering & Prompt Engineering!
🐰🎙️
A CodeRabbit review per day, keeps the Merge Cop away!
November 10, 2025 at 4:00 PM
A CodeRabbit review per day, keeps the Merge Cop away!
The problem with 👍👎 is they optimize LLMs for approval, not truth.
Stop training dev tools to chase treats like golden retrievers. Give high-signal, contextual feedback instead.
Heres how!
Stop training dev tools to chase treats like golden retrievers. Give high-signal, contextual feedback instead.
Heres how!
November 7, 2025 at 4:27 PM
The problem with 👍👎 is they optimize LLMs for approval, not truth.
Stop training dev tools to chase treats like golden retrievers. Give high-signal, contextual feedback instead.
Heres how!
Stop training dev tools to chase treats like golden retrievers. Give high-signal, contextual feedback instead.
Heres how!