Govind Ramnarayan
Senior Deep Learning Software Engineer at NVIDIA
Seattle, Washington, United States
Summary
🤩
Rockstar🎓
Top SchoolGovind Ramnarayan is a Senior Deep Learning Software Engineer in Seattle with five years of focused experience building high-performance ML systems and runtimes. He has driven production-ready optimizations for sparse neural network inference at Neural Magic and advanced secure computation and enclave backends at Pyte, bridging research-quality algorithms with customer-facing products. His recent work includes LLM data engineering for a stealth startup and now contributing to deep learning software at NVIDIA, reflecting a blend of systems-level performance tuning and applied ML research. Trained at UC Berkeley and pursuing a PhD at MIT in computer science, he pairs rigorous academic background with practical engineering outcomes. Notably, he has deep expertise in memory- and cache-aware techniques (e.g., Tensor Columns) that materially speed CPU-based neural workloads—skills that translate into faster model deployment and lower infrastructure costs.
5 years of coding experience
5 years of employment as a software developer
Bachelor of Science (B.S.) Electrical Engineering and Computer Science, Bachelor of Science (B.S.) Electrical Engineering and Computer Science at University of California, BerkeleyDoctor of Philosophy - PhD Computer Science, Doctor of Philosophy - PhD Computer Science at Massachusetts Institute of Technology
Github Skills (17)
pruning10
transfer-learning10
natural-language-processing9
keras9
nlp9
object-detection9
computer-vision8
quantization8
automl8
onnx8
pytorch8
image-classification7
llm-inference7
inference7
tensorflow7
Programming languages (1)
Github contributions (2)
neuralmagic/deepsparse
Apr 2022 - Apr 2022
Sparsity-aware deep learning inference runtime for CPUs
Contributions:2 reviews, 1 commit, 2 PRs in 1 day
deep-learninginferencemachinelearningonnxcomputer-vision
neuralmagic/sparseml
Feb 2021 - Mar 2021
Libraries for applying sparsification recipes to neural networks with a few lines of code, enabling faster and smaller models
Contributions:3 reviews in 1 month
neural-networkpytorchkerassparsification-recipestensorflow
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.