ML Engineering Manager - Distributed Inference at Apple
San Francisco Bay Area United States
Join Prog.AI to see contacts
Join Prog.AI to see contacts
Summary
🤩
Rockstar
🎓
Top School
Denis-gabriel Vieriu is a Senior Software Engineer based in the San Francisco Bay Area, specializing in machine learning and GPU software for Apple silicon. With a background that spans hypervisor engineering at Bitdefender to GPU driver and Metal feature development at Apple, he blends low-level systems expertise with ML-focused engineering. He has directly contributed to PyTorch’s MPS backend, improving performance and correctness for Apple GPUs—a notable open-source impact on a widely used ML framework. At Apple he designed and implemented features such as Mesh Shader Emulation and Function Pointers, demonstrating a knack for bridging kernel/user-mode concerns with high-performance compute. Trained in mathematics and computer science, he brings rigorous problem-solving and a systems-first mindset to production ML and graphics stacks. Colleagues rely on him to turn subtle memory/layout and backend inconsistencies into robust, high-performance implementations.
4 years of coding experience
10 years of employment as a software developer
Bachelor's degree Computer Science, Bachelor's degree Computer Science at Babes-Bolyai University, Faculty of Mathematics and Computer Science
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Role in this project:
ML Engineer
Contributions:94 reviews, 30 commits, 78 PRs in 6 months
Contributions summary:Denis-gabriel primarily contributed to the PyTorch framework by implementing and fixing operations related to the MPS (Metal Performance Shaders) backend, focusing on enabling and optimizing machine learning computations on Apple silicon. Their work included adding test cases, handling boolean inputs, fixing placeholder cases, fixing memory casts, fixing consistency errors, adding support for various ops, and addressing issues related to view tensors and contiguous memory layouts. The user's contributions directly impacted the performance and functionality of PyTorch on Apple devices.
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Contributions:212 reviews, 472 commits, 231 PRs in 8 months
gpu-accelerationneural-networkpython
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.