Repo Roast – a real-time feed of the funniest comments on GitHub
Hey HN - I'm Mason, and I'm excited to share a fun side project I've been working on recently: Repo Roast. For me, Github often feels like a serious (and sometimes soulless) place. I built this website to showcase the human, silly side of developers. Repo Roast is built on top of LLM-powered GitHub analytics tooling we’re developing internally at Merit Systems [0]. One of the historically cited weak points for LLMs has been their sense of humor. I found that with the right prompt, at scale, they mostly can _identify_ humor pretty well. We experimented with a couple of different models to…
In plain words
Repo Roast aggregates funny comments from GitHub repositories and displays them in a real-time feed. Built using AI to identify humorous posts at scale, the tool uses a combination of language models to filter and score comments for comedic value. It's designed for developers who want to see the lighthearted and human side of GitHub, which is often perceived as a serious platform. The project showcases how AI can effectively identify humor when given the right prompts and approach.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
Hey HN - I'm Mason, and I'm excited to share a fun side project I've been working on recently: Repo Roast. For me, Github often feels like a serious (and sometimes soulless) place. I built this website to showcase the human, silly side of developers. Repo Roast is built on top of LLM-powered GitHub analytics tooling we’re developing internally at Merit Systems [0]. One of the historically cited weak points for LLMs has been their sense of humor. I found that with the right prompt, at scale, they mostly can _identify_ humor pretty well. We experimented with a couple of different models to tune the cost/effectiveness and eventually landed on a strategy where we have gpt-4o-mini do a very light filter pass to reduce the number of calls to GPT-4o to do the actual humor scoring. Interestingly, our first prompt attempts ended up surfacing wayyy more aggressively NSFW content, which is funny that this happens despite all the censoring that OpenAI is doing. So far we have processed the last 90 days of historical comments and continuously process new ones every hour. We theoretically can run this against a wider historical window given there is sufficient interest to justify the costs! Repo Roast is in no way intended to make you a better or more productive dev. You can use it as a leaderboard or hiring tool (if you’re looking for personality), or just a way to distract yourself! If you've had any funny Github interactions, I'd love if you shared them below. Cheers! --Mason [0] https://www.merit.systems/
Does the same job
all alternatives →
RoastMyGitApr 2026 · roastmygithub.lovable.app · ▲2Get a hilarious AI roast based on your GitHub stats
GitHub Roast Jul 2026 · gitroast-kappa.vercel.app · ▲2Brutally honest AI feedback for your GitHub profile
- RARoastedby.ai – Talk some trash, have some fun2023 · roastedby.ai · ▲46
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 27d ago · cactuscompute.com
Launched alongside, April 2025
the whole month →- IB
Hi everyone, I built PyXL — a hardware processor that executes a custom assembly generated from Python programs, without using a traditional interpreter or virtual machine. It compiles Python -> CPython Bytecode -> Instruction set designed for direct hardware execution. I’m sharing an early benchmark: a GPIO test where PyXL achieves a 480ns round-trip toggle — compared to 14-25 micro seconds on a MicroPython Pyboard - even though PyXL runs at a lower clock (100MHz vs. 168MHz). The design is stack-based, fully pipelined, and preserves Python's dynamic typing without static type restrictions.…
Dev tools · 2025 · runpyxl.com
- UC
Life & fun · 2025 · filiph.github.io
- IB
https://the-pocket.github.io/Tutorial-Codebase-Knowledge/
AI · 2025 · github.com
