Reiner Pope is Co-Founder and CEO of MatX, a semiconductor company designing specialized chips for large language models to deliver 10x more computing power for AI labs. MatX aims to make frontier AI models run as efficiently as physics allows, bringing advanced AI capabilities years ahead in quality and availability.

Reiner holds a Bachelor of Philosophy (Honours) from The Australian National University. Prior to founding MatX in 2022, he spent over a decade at Google, progressing from Software Engineer to Senior Staff Software Engineer at Google Research, where he worked on LLM hardware and software efficiency, compilers, and chip architecture.

Reiner was a core contributor to Google PaLM, the 540-billion parameter language model, and built the world's highest-performing LLM inference software. He also worked at X: The Moonshot Factory as a Software Engineer and Architect, and has authored highly-cited research papers with over 11,500 citations.

MatX raised $500M in Series B funding in February 2026, led by Jane Street and Situational Awareness, valuing the company at over $300M in its 2024 Series A. The company has grown to about 100 employees and is partnering with TSMC to manufacture chips, targeting shipment in 2027.

MatX is developing the MatX One chip that combines HBM and SRAM architectures, claiming 10x better efficiency than Nvidia's upcoming Rubin Ultra. The company's hardware is designed to train models like GPT-4 and run ChatGPT-scale applications on a small startup budget.