Atharva Dubey
Senior Machine Learning Engineer at Synthesia
City of Edinburgh, Scotland, United Kingdom
Summary
🤩
Rockstar🎓
Top SchoolAtharva Dubey is a Software Engineer 2 with six years of experience specializing in high-performance GPU computing, primarily using SYCL and CUDA to optimize algorithms and compute libraries. At Codeplay he drives SYCL backend development for oneDNN, prototyped a template-heavy DNN library with runtime compilation for kernel specialization and fusion, and led performance work on llama.cpp inference across Intel GPUs. His background includes low-level operator implementation and quantized neural-net optimizations, giving him a strong cross-stack view from framework to hardware ISA. Based in Edinburgh, he combines applied compiler work with practical engineering—extending libraries like CUTLASS to new hardware—and is continuously learning compiler internals while contributing to open-source GPU projects.
6 years of coding experience
4 years of employment as a software developer
BITS Pilani, Birla Institute of Technology and Science
Stackoverflow
862reputation
121kreached
8answers
40questions
Github Skills (61)
intel10
python10
image10
image-processing10
llvm10
machine-learning10
llm-inference10
numpy10
sycl10
openmp10
tbb10
oneapi10
deep-learning10
gpu10
x86-6410
Programming languages (6)
Github contributions (5)
AD2605/oneDNN
Oct 2024 - Jun 2025
oneAPI Deep Neural Network Library (oneDNN)
Contributions:1 PR, 106 pushes, 9 branches in 7 months
AD2605/llvm
Apr 2024 - Aug 2025
Intel staging area for llvm.org contribution. Home for Intel LLVM-based projects.
Contributions:2 reviews, 2 PRs, 124 pushes in 1 year 4 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.