Jagrit Digani
Machine Learning Engineer at Apple
San Jose, California, United States
Summary
🤩
Rockstar🎓
Top SchoolJagrit Digani is a Machine Learning Engineer based in San Jose with six years of hands-on experience applying deep learning to computer vision and inverse-design problems. Currently at Apple, he focuses on performance-sensitive ML engineering and has contributed low-level optimizations to an Apple-silicon array framework (MLX), including new kernels and GEMM improvements. His background blends academic research—developing transformer-based inverse design models and studying black-box optimization—with product-focused work like mobile-optimized CNNs for sports analytics and pose estimation. He brings practical systems skills (CUDA, Docker, CoreML, TorchScript) and a track record of accelerating simulations and training pipelines by up to four times. Curious and pragmatic, he often bridges physics-driven simulation and ML, producing reproducible toolchains and synthetic data pipelines that make research solutions production-ready.
6 years of coding experience
3 years of employment as a software developer
Bachelor of Science - BS, Computer Science, Bachelor of Science - BS, Computer Science at University of California, Los Angeles
Github Skills (12)
kernel10
metal10
matrix-multiplication10
gemfire10
c-language10
gpgpu10
cprogramming-language10
gpu10
build-system7
gnu-make7
cmake7
documentation5
Programming languages (2)
Github contributions (5)
ml-explore/mlx
Dec 2023 - May 2026
MLX: An array framework for Apple silicon
Role in this project:
Back-end Developer
Back-end DeveloperContributions:2 releases, 112 reviews, 92 PRs in 2 years 5 months
Contributions summary:Jagrit contributed significantly to the MLX array framework for Apple silicon, focusing on low-level optimizations and new feature implementations. Their work involved implementing new kernels, optimizing existing GEMM (matrix multiplication) routines, and addressing a broadcasting bug. Furthermore, they enhanced the framework with a new SliceUpdate operation and primitive, demonstrating a focus on performance and efficiency. The user also integrated updates to the build process and documentation.
apple-siliconmlx
Upsilon-Wave-Team/Physics-Engine
Jan 2020 - Nov 2020
Contributions:21 pushes, 1 branch in 9 months
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.