# Reiner Pope

> Reiner Pope is the co-founder and CEO of MatX, an AI chip startup based in Mountain View, California, designing purpose-built silicon for large language models. A former Google Senior Staff Software Engineer who led AI software development for Google's TPUs and served as Efficiency Lead for PaLM, Pope left Google one week before ChatGPT launched to build what he calls 'the best chips for LLMs that physics allows.' MatX has raised approximately $604M to date, including a $500M Series B led by Jane Street and Situational Awareness LP, and is building a hybrid SRAM-HBM chip architecture targeting the world's leading AI labs.

- **Role:** Founder and CEO at MatX
- **Organizations:** MatX, Google Research, Google Brain
- **Nationality:** Australian
- **Education:** Bachelor of Philosophy (Honours) in Mathematics, The Australian National University
- **Known for:** Co-authored PaLM: Scaling Language Modeling with Pathways (2023) - 9,400+ Google Scholar citations, Co-authored PaLM 2 Technical Report (2023) - 2,459+ citations, Led 'Efficiently Scaling Transformer Inference' paper (2023) - 919+ citations

## Career timeline

- **2011** — Graduated from The Australian National University with a BPhil (Hons) in Mathematics
- **2013-2019** — Software engineer roles focused on compilers, algorithms, and systems; developed expertise in Haskell and low-level optimization
- **2019** — Joined Google Research as Senior Staff Software Engineer, working on LLM hardware/software efficiency, compilers, and TPU architecture
- **2020-2022** — Served as Efficiency Lead for Google PaLM; designed and implemented world's fastest LLM inference software; helped conceive Google TPU v5e
- **2022** — Co-authored foundational PaLM paper (9,400+ citations) and 'Efficiently Scaling Transformer Inference' (919 citations)
- **2022-11** — Left Google one week before ChatGPT launched to co-found MatX with Mike Gunter (former lead TPU hardware designer)
- **2023** — Founded MatX; raised Series A of approximately $100M at over $300M valuation
- **2025** — Published blog posts on algorithmic optimization: cuckoo hashing, Strassen matmuls for AI, neural network/cipher analogies
- **2026-02** — MatX closed $500M Series B led by Jane Street and Situational Awareness LP, total funding reaching ~$604M

## Achievements

- Co-authored PaLM: Scaling Language Modeling with Pathways (2023) - 9,400+ Google Scholar citations
- Co-authored PaLM 2 Technical Report (2023) - 2,459+ citations
- Led 'Efficiently Scaling Transformer Inference' paper (2023) - 919+ citations
- Designed and implemented Google's world-fastest LLM inference software as Efficiency Lead for PaLM
- Helped conceive Google's TPU v5e chip architecture
- Co-founded MatX, which raised $604M total including $500M Series B from Jane Street and Situational Awareness LP
- Built MatX to 110+ employees within two years of founding
- Co-authored JAX scaling book available at jax-ml.github.io/scaling-book/
- Co-authored SPIRe: Boosting LLM Inference Throughput with Speculative Decoding (2025)

## Latest updates

- **2026-02** — MatX closed $500M Series B funding led by Jane Street and Situational Awareness LP, with participation from Marvell Technology, Spark Capital, NFDG, and Stripe co-founders Patrick and John Collison. Total funding now ~$604M.
- **2025-04** — Co-authored SPIRe: Boosting LLM Inference Throughput with Speculative Decoding, published on arXiv.
- **2025** — Published multiple technical blog posts on reiner.org covering algorithmic research including cuckoo hashing for SIMD hash tables, hashed sorting vs. hash tables, Strassen matmuls and AI, and neural network/cipher structural analogies.

## Links

- Website: https://reiner.org/
- LinkedIn: http://www.linkedin.com/in/reiner-pope-08064345
- Twitter/X: https://x.com/reinerpope

---

Profile page: https://yespress.io/reiner-pope
Published by YesPress — https://yespress.io
Last updated: 2026-05-23