Ankur Neog
Software Engineer at Intel Corporation
Bengaluru, Karnataka, India
Summary
🤩
Rockstar🎓
Top SchoolAnkur Neog is a software engineer with 8 years of experience specializing in AI software and systems, currently advancing Intel’s AI platforms from Bengaluru. He brings deep device-level ML engineering expertise, notably contributing PyTorch support and testing for Intel Gaudi/HPU devices to enable reliable GPU/accelerator integration. His career spans engineering and principal roles at Infinera, Juniper, and Flex, reflecting strong systems and firmware-to-software breadth across networking and compute domains. Ankur pairs hands-on open-source work with production-grade platform delivery, and his contributions to one of the most prominent ML frameworks underline a practical focus on making cutting-edge hardware usable by mainstream ML stacks.
8 years of coding experience
18 years of employment as a software developer
BITS Pilani, Birla Institute of Technology and ScienceBachelor of Engineering - BE, Bachelor of Engineering - BE at National Institute of Technology Silchar
Github Skills (10)
pytorch10
machine-learning10
deeplearning-ai10
deep-learning10
gpu10
python10
test-automation10
neural-network9
tensor9
autograd8
Programming languages (1)
Github contributions (5)
pytorch/pytorch
May 2024 - Mar 2025
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Role in this project:
ML Engineer
ML EngineerContributions:32 reviews, 14 PRs, 201 comments in 10 months
Contributions summary:Ankur primarily contributes to integrating and testing the PyTorch framework with Intel Gaudi (HPU) devices. Their work involves adding HPU device support within the testing framework, including modifications to common device type tests and utilities. The contributions also include adding dtype exceptions, extending decorators, and generalizing device-specific tests for the Dynamo backend. Furthermore, the user is working on adding support for Gaudi/HPU in the FSDP state dict utilities, and adding dtype support to the opinfo to include the HPU devices.
pythongpu-accelerationdeep-learninggpunumpy
ankurneog/pytorch
May 2024 - Mar 2025
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Contributions:94 pushes, 21 branches 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.