SOURCE-LINKED INTELLIGENCE

MaxKernel: Agentic Kernel Generation for TPUs

arXiv · AI, language, vision and robotics · article · Sep 3, 2026 · UTC

Designing and authoring high-performance custom kernels for accelerators is a complex task that requires deep hardware-level expertise. Large Language Models (LLM) can be leveraged together with real-time compiler feedback to build agentic systems for kernel generation. In this work, we present MaxKernel, a multi-agent system that implements three distinct paradigms for TPU kernel development: (1) a Human-in-the-Loop (HITL) agent for collaborative, step-by-step design; (2) an Autonomous (Auto) agent that executes a fully automated, metric/trace-driven optimization loop; and (3) a Graph-Based A

Read original source ↗ Open in workspace

recordType

paper

region

Global