Lessons From Training Composer At Cursor And Building Meta/Nvidia Compute Clusters | YC Paper Club
At our latest YC Paper Club, researchers and builders presented on multi-GPU kernel optimization, intelligence per watt for local inference, AI-generated GPU kernels and benchmarking, heterogeneous inference infrastructure design, and GPU-accelerated game engines for reinforcement learning. Thanks to the following presenters:
Stuart Sul (Stanford / Cursor), John (Stanford), Mark (PyTorch / GPU Mode / CoreAuto), Misha (Marlo), and Brennan (Stanford)
Transcript: https://www.ycrootaccess.com/p/multi-...
Chapters:
0:00 – Francois Chaubard: The case for chip and kernel specialization
7:16 – Stuart Sul: Parallel Kittens - Systematic and Practical Simplification of Multi-GPU Al Kernels (https://arxiv.org/abs/2511.13940)
21:29 – Jon Saad-Falcon: Intelligence per Watt - Measuring the Intelligence Efficiency of Local and Cloud AI (https://arxiv.org/abs/2511.07885)
31:05 – Mark Saroufim: When Al Starts Writing Systems Code
47:04 – Misha Smelyanskiy: Why AI Inference Needs Heterogeneous Hardware
1:04:33 – Brennan Shacklett: Building a High-Throughput Game Engine that Runs ENTIRELY on the GPU (https://madrona-engine.github.io/shac...)
1:15:35 – Wrap-up & what's next
Video Summary
AI GeneratedThis session of the YC Paper Club explores the specialization of AI hardware and software, focusing on GPU kernel optimization, the shift toward local inference, and the use of heterogeneous architectures to resolve bottlenecks in LLM prefill and decode phases. It also highlights the emergence of AI-generated kernels and high-throughput GPU-based simulation engines.
