# Jonathan Summitt Jonathan Summitt is the co-founder and CEO of Groq, an AI chip company. He previously led the development of Google's Tensor Processing Unit (TPU), a custom ASIC for machine learning. Jonathan Summitt is an American technology executive and computer engineer. He is the co-founder and chief executive officer (CEO) of Groq, a company that designs application-specific integrated circuits (ASICs) for artificial intelligence (AI) workloads. Summitt is best known for his prior role as the lead of the Tensor Processing Unit (TPU) project at Google, where he oversaw the development of custom hardware that accelerated [machine-learning](https://www.wikiprompt.org/wiki/machine-learning) tasks within the company's data centers. Summitt's career has centered on the intersection of hardware architecture and [artificial-intelligence](https://www.wikiprompt.org/wiki/artificial-intelligence). His work at Google and later at Groq has focused on creating specialized processors that offer higher performance and efficiency for [deep-learning](https://www.wikiprompt.org/wiki/deep-learning) models compared to general-purpose graphics processing units (GPUs). He has been a prominent figure in the movement toward domain-specific computing for AI. ## Early Career and Education Summitt earned a bachelor's degree in electrical engineering from the University of Texas at Austin in 1994. He subsequently worked at several semiconductor companies, including Texas Instruments and Sun Microsystems, where he gained experience in microprocessor design. In the early 2000s, he joined a startup called Stream Processors, Inc., which focused on stream processing architectures, a field related to dataflow computing that would later influence his work at Groq. ## Google TPU Leadership Summitt joined Google in 2013 as a hardware engineer. He was soon tasked with leading the development of the TPU, a custom ASIC designed to accelerate inference for [neural-network](https://www.wikiprompt.org/wiki/neural-network) models, particularly those used in search ranking and [large-language-model](https://www.wikiprompt.org/wiki/large-language-model) applications. The first-generation TPU was deployed in Google's data centers in 2015 and was publicly announced in 2016. Summitt's team later developed the TPU v2 and v3, which added training capabilities and were used for projects such as AlphaGo and Google Translate. During his tenure at Google, Summitt also contributed to the broader [google-cloud](https://www.wikiprompt.org/wiki/google-cloud) strategy, making TPUs available to external customers through cloud services. He left Google in 2016 to co-found Groq, though he remained involved in the TPU ecosystem as an advisor for a period. ## Groq and the LPU In 2016, Summitt co-founded Groq with a team of former Google TPU engineers. The company's flagship product is the Language Processing Unit (LPU), a processor designed specifically for the [transformer](https://www.wikiprompt.org/wiki/transformer) architecture that underpins many modern [generative-ai](https://www.wikiprompt.org/wiki/generative-ai) systems. The LPU uses a dataflow architecture, where instructions are scheduled statically at compile time rather than dynamically at runtime, reducing overhead and improving latency. Groq's first LPU chip, the GroqChip, was announced in 2020 and began shipping in 2021. In 2024, Groq launched a cloud service that provides access to LPUs for running [large-language-model](https://www.wikiprompt.org/wiki/large-language-model) inference. The service gained attention for its high token-generation speeds, which Summitt has publicly demonstrated in benchmark comparisons against GPU-based systems from [nvidia](https://www.wikiprompt.org/wiki/nvidia) and [amd](https://www.wikiprompt.org/wiki/amd). As of 2025, Groq has raised over $1 billion in funding from investors including [intel](https://www.wikiprompt.org/wiki/intel) Capital and [samsung-electronics](https://www.wikiprompt.org/wiki/samsung-electronics). ## Industry Influence and Recognition Summitt has been a vocal advocate for custom silicon in AI, arguing that general-purpose GPUs are not optimal for the specific demands of [deep-learning](https://www.wikiprompt.org/wiki/deep-learning) inference. He has spoken at major conferences such as the International Solid-State Circuits Conference (ISSCC) and the Hot Chips symposium. In 2023, he was named a Fellow of the Institute of Electrical and Electronics Engineers (IEEE) for contributions to the design of tensor processing units. His work has also influenced other AI hardware startups, including [samba-nova](https://www.wikiprompt.org/wiki/samba-nova) and [graphcore](https://www.wikiprompt.org/wiki/graphcore), which have adopted similar approaches to domain-specific architecture. Summitt's leadership at Groq has positioned the company as a key player in the competitive AI chip market, alongside established firms and cloud providers. ## Personal Life and Public Profile Summitt is known for his hands-on engineering approach and often participates in technical discussions on social media. He has been based in the San Francisco Bay Area since joining Google. Details about his personal life are limited, as he tends to keep a low profile outside of professional engagements. He has occasionally published technical blog posts about Groq's architecture and the future of AI hardware. ## See Also - [google-deepmind](https://www.wikiprompt.org/wiki/google-deepmind) - [azure](https://www.wikiprompt.org/wiki/azure) - [amazon-web-services](https://www.wikiprompt.org/wiki/amazon-web-services) - [oracle-cloud](https://www.wikiprompt.org/wiki/oracle-cloud) - [openai](https://www.wikiprompt.org/wiki/openai) - [anthropic](https://www.wikiprompt.org/wiki/anthropic) --- Source: https://www.wikiprompt.org/wiki/jonathan-summit License: CC BY-SA 4.0 (https://creativecommons.org/licenses/by-sa/4.0/) Last updated: 2026-09-12T22:25:25.435124+00:00