Show HN: A CUDA-Generating AI Agent

Hacker News - AI
Jul 9, 2025 23:02
atallahw
1 views
hackernewsaidiscussion

Summary

MakoGenerate is an AI agent that automatically converts PyTorch code into optimized CUDA kernels, iteratively improving performance and correcting errors based on feedback from compilation and validation. Users can customize the process with prompts, documentation, and specific optimization goals, making it a flexible tool for accelerating deep learning workflows. This approach has significant implications for automating and streamlining the development of high-performance AI models on GPU hardware.

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 configure the agent with a model/hardware/compute budget. The optional custom prompt section gives you the opportunity to add more context to the agent. You can include specific documentation, tutorials, or examples. You can also identify specific optimization techniques that you would like to see applied. Prompts can be as short as one line, or even multiple pages long. Let me know what you think. Comments URL: https://news.ycombinator.com/item?id=44515598 Points: 2 # Comments: 0

Related Articles

Kiro vs. Cursor – AI IDE comparison breakdown

Hacker News - AIJul 17

The article compares Kiro and Cursor, two AI-powered integrated development environments (IDEs), highlighting their respective strengths in code generation, user interface, and collaboration features. It notes that while both tools leverage advanced AI to boost developer productivity, their differing approaches cater to distinct user needs. The comparison underscores the growing impact of AI-driven IDEs on software development workflows.

How Artificial Intelligence is Driving the Future of Biocomputing?

Analytics InsightJul 17

The article discusses how artificial intelligence is accelerating advancements in biocomputing by enabling more efficient analysis of complex biological data and the design of novel biomolecular systems. It highlights AI’s role in breakthroughs such as protein folding prediction and synthetic biology, suggesting that this synergy could revolutionize healthcare, drug discovery, and computational biology. The integration of AI and biocomputing is poised to drive innovation and reshape the future of both fields.

Hate Windows 11? Here's how you can make it work more like Windows 10

ZDNet - Artificial IntelligenceJul 17

The article discusses user frustrations with Windows 11 and offers solutions to make its interface resemble Windows 10. While not directly focused on AI, these customization options may improve user experience and productivity, which could indirectly benefit AI workflows and adoption on Windows platforms.