matx

MatX builds custom silicon and vertically integrated full-stack solutions to train and run the world's largest AI models, combining hardware and software co-design across instruction set architectures, register-transfer level design, simulators, compilers, and kernels. The company serves users requiring efficient generative artificial intelligence infrastructure, aiming to make advanced AI models run efficiently and accelerate AI quality and availability. Its stated work in artificial intelligence and machine learning involves developing custom silicon for high-performance and sustainable generative AI, defining architectures for compute subsystems, and applying coding agents to hardware engineering workflows.

The open roles indicate that MatX is building multidisciplinary teams across architecture, infrastructure, and software engineering. The available positions include AI Chip Architect, Infrastructure Engineer for AI Tooling, Software Engineer for AI Tooling, and Performance Modeling Engineer. These roles focus on AI tooling, evaluation loops, agent knowledge bases, infrastructure, and chip architecture. The open positions are primarily located at the company headquarters in Mountain View, California, with hybrid expectations or remote work availability depending on the specific role.