atallahw
Re:RL on GPT-5 to write better kernels
on Thursday February 12, 2026 @12:22PM (#46991735)
We got access to OpenAI's RFT API with GPT5 and tried to see how good we could get it at one-shot Triton kernel generation. Some key decisions/observations:
1. tool use instead of multi-turn rl
2. skip SFT altogether
3. dataset curation was more important than dataset scale
4. reward hacks detection must be robust
5. models are getting a lot better at this
Re:Scammed out of $130K via fake Google call, spoofed Google email and...
on Tuesday September 16, 2025 @01:30PM (#45265219)
What did the account did the email actually come from? Was it legit from legal and he just submitted the request or was it a real spoofing
Show HN: A CUDA-Generating AI Agent
MakoGenerate is an AI agent that takes any PyTorch code as input, and writes CUDA kernels for it.You can choose a sample problem or enter in your own PyTorch. It generates CUDA, compiles and validates the results, then uses all of this as feedback into the next generation attempt. If the generated kernel didn’t work, it tries to fix it. If it did work, it tries to make it faster.You can...
Re:MakoGenerate: AI-Powered GPU Kernel Generation in Under 60 Seconds
on Wednesday June 25, 2025 @11:17AM (#44378317)
This was fun to work on. LLMs for writing kernels still has a long way to go. Its honestly a little surprising how decent they are now. I guess I've been pretty consistently "surprised" by codegen for a while now (meaning the last two years)