Summary
Suvinay Subramanian is a software engineer at Google with 13 years of experience focused on designing and optimizing specialized hardware-accelerator systems (TPUs) for AI workloads like LLMs and recommendation models. He brings deep hardware-software codesign expertise spanning computer architecture, parallel programming, distributed systems, and networking, grounded in an MS and PhD from MIT and a BTech from IIT Madras. At Google since 2018, he translates research-grade architectural ideas into high-performance production systems that improve model throughput and efficiency at scale. Known for bridging low-level systems thinking with practical software delivery, he excels at making complex co-design tradeoffs measurable and repeatable. Based in San Francisco, he combines academic rigor with hands-on engineering to tackle bottlenecks in modern AI infrastructure. Less obvious: his career blends long-term research training with product-focused execution, enabling rapid iteration across both hardware and software layers.
13 years of coding experience
Indian Institute of Technology Madras
Master of Science - MS, Electrical Engineering and Computer Science, Master of Science - MS, Electrical Engineering and Computer Science at Massachusetts Institute of Technology