avatar

Matt Suiche

Hacker · Founder of OnDB

2026

The 1.2 ms Eigensolver That Never Ran
GPU MODE NVIDIA B200 CUDA cuSOLVER PyTorch Triton Eigendecomposition Benchmarking Reward Hacking AI Agents

2025

Optimizing AlphaFold's Triangle Multiplicative Update: A First Look at GPU Performance Engineering
GPU Optimization PyTorch Triton AlphaFold Machine Learning Performance Engineering H100 Tensor Cores
Gluon: When Triton Isn't Low-Level Enough
GPU Triton Gluon Performance CUDA Deep Learning PyTorch
The Hidden Math Bug That Makes AI Unpredictable
determinism floating-point neural-networks pytorch mlx