Abhinav Goel
Senior Deep Learning Architect at NVIDIA
California, United States
Summary
👤
Senior🎓
Top SchoolAbhinav Goel is a Senior Deep Learning Architect based in California with a decade of experience accelerating AI models on GPUs and shaping training performance for JAX at NVIDIA. He blends deep academic rigor from a PhD in Computer Engineering with hands-on GPU performance engineering and computer architecture expertise, having led technical teams focused on making state-of-the-art models run faster. His background includes low-power computer vision research for edge devices, multiple performance internships at Qualcomm that yielded granted patents, and contributions to the widely used JAX ecosystem—authoring tooling to convert NVIDIA Nsys profiles for XLA’s latency estimator. Known for pragmatic, quality-first code, he bridges profiling, compiler-aware optimization, and architecture-level reasoning to squeeze latency and efficiency gains out of real systems.
9 years of coding experience
4 years of employment as a software developer
Bachelor of Technology (B.Tech.), Electronics and Communication Engineering, Bachelor of Technology (B.Tech.), Electronics and Communication Engineering at PES UniversityDoctor of Philosophy - PhD, Computer Engineering, Doctor of Philosophy - PhD, Computer Engineering at Purdue University
English, Hindi
Github Skills (5)
xla10
python10
profiling10
jax9
mlops7
Programming languages (7)
Github contributions (5)
jax-ml/jax
Nov 2022 - Oct 2024
Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more
Role in this project:
ML Engineer
ML EngineerContributions:1 review, 3 PRs, 8 comments in 1 year 10 months
Contributions summary:Abhinav contributed to a tool for converting NVIDIA Nsys profiles to the .pbtxt format, specifically for use with XLA's Profile Guided Latency Estimator. Their work involved writing a Python script that parses the profiling data and extracts timing information for different operations. The commits include adding license information, responding to reviewer comments, and updating the script to handle different report formats. The user also fixed type-related issues, indicating a focus on code quality and maintainability within the project.
pytorchpythonjitautomatic-differentiationgpu
abhinavgoel95/TRUNK
Feb 2022 - Feb 2022
Contributions:4 commits, 3 pushes, 1 branch in 1 day
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.