AI RESEARCH
AutoKernel: Autonomous GPU Kernel Optimization via Iterative Agent-Driven Search
arXiv CS.LG
•
ArXi:2603.21331v1 Announce Type: new Writing high-performance GPU kernels is among the most labor-intensive tasks in machine learning systems engineering. We present AutoKernel, an open-source framework that applies an autonomous agent loop to GPU kernel optimization for arbitrary PyTorch models. Given a model, AutoKernel profiles it to identify computational bottlenecks, ranks them by Amdahl's law impact, and iteratively refines Triton or CUDA C++ kernel implementations through hundreds of experiments without human intervention.