Kabir Nagrecha
Chief Executive Officer at Tessera Labs
San Jose, California, United States
Summary
👤
Senior🎓
Top SchoolKabir Nagrecha is a founder-CEO and machine learning systems expert with eight years of experience building and optimizing large-scale deep learning infrastructure for industry leaders like Meta and Netflix. He earned a Ph.D. from UC San Diego at an unusually young age and has focused his career on model-system co-design, performance optimizations, and productionizing LLMs and diffusion models for recommendations and content generation. At Meta he worked on efficiency for ads and LLaMA-class models, and at Netflix he accelerated studio and recs workflows—experience he now channels into Tessera Labs to transform enterprise operations with GenAI. Known for tackling deep technical problems with broad scope, he blends research rigor with product and engineering execution. He has a track record of dramatic runtime and error-rate improvements from prior internships (e.g., Siri pipelines at Apple) and is actively hiring across product, AI, and GTM as his startup emerges from stealth. Open to conversations on research and ambitious AI products, Kabir combines academic depth with hands-on systems impact.
8 years of coding experience
6 years of employment as a software developer
University of California, San DiegoDoctor of Philosophy - PhD Computer Science, Doctor of Philosophy - PhD Computer Science at UC San Diego Jacobs School of EngineeringBachelor of Science - BS Computer Science, Bachelor of Science - BS Computer Science at California State University, Los Angeles
Stackoverflow
13reputation
692reached
0answers
1question
Github Skills (63)
python10
acceleration10
machine-learning10
numpy10
deep-learning10
tensorflow10
gpu10
autograd10
neural-network10
tensor10
gpu-acceleration10
parallelization9
pytorch9
mixture-of-experts9
query-optimization9
Programming languages (5)
Github contributions (5)
knagrecha/hydra
Jul 2021 - Nov 2022
Execution framework for multi-task model parallelism. Enables the training of arbitrarily large models with a single GPU, with linear speedups for multi-gpu multi-task execution.
Contributions:241 commits, 3 PRs, 273 pushes in 1 year 4 months
traininglarge-modelsmulti-gpumulti-taskdistributed-training
StevenHTWu/TritonEats
Nov 2020 - Dec 2020
Contributions:4 reviews, 13 PRs, 51 pushes in 27 days
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.