Siva Reddy
Engineer, Senior Staff at The Apache Software Foundation
Bengaluru, Karnataka, India
Summary
🤩
Rockstar🎓
Top SchoolSiva Reddy is a Senior Staff Engineer at Qualcomm with two decades of industry experience focused on product development, platform technologies, and systems programming across Linux, Android, and embedded GPUs. He blends deep expertise in AI and deep learning compilers with hands-on back-end development, contributing notable fixes and features to high-profile open-source projects like TVM and NNVM for ONNX compatibility and graph/debug tooling. As an Apache committer, he has demonstrated strengths in shape inference, operator implementation, and optimization for GPU and CPU targets, bridging research and production. Based in Bengaluru, he pairs formal study in AI/ML from Illinois Institute of Technology with practical embedded systems roots, dating back to early Linux work. Less obvious: he routinely surfaces hard-to-find graph-debug issues and layout-conversion bugs that improve model portability across accelerators.
8 years of coding experience
2 years of employment as a software developer
Jawaharlal Nehru Technological University HyderabadPGP - Artificial Intelligence & Machine Learning, Information Technology, PGP - Artificial Intelligence & Machine Learning, Information Technology at Illinois Institute of Technology
Github Skills (16)
compiler10
tvm10
nvm10
machine-learning10
relay-modern10
compiler-compiler10
deep-learning10
tensorflow10
onnx10
python10
relay10
c-language8
cprogramming-language8
gpu8
optimization7
Programming languages (5)
Github contributions (5)
apache/tvm
Nov 2017 - Jan 2023
Open deep learning compiler stack for cpu, gpu and specialized accelerators
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:136 reviews, 92 commits, 174 PRs in 5 years 3 months
Contributions summary:Siva's commits primarily focus on enhancing the "get_started" tutorial script, adding debug functionality to retrieve node outputs, and fixing graph debug issues. They also made corrections to function headers and shape inference for the "NNVM" and ONNX frontends, specifically relating to reshape, squeeze and other tensor operations. The user demonstrated skills related to compiler development, specifically the TVM compiler stack, by integrating and adapting it for various models.
metalvulkancompilertensoropencl
dmlc/nnvm
Nov 2017 - May 2018
Role in this project:ML Engineer
ML EngineerContributions:11 commits, 20 PRs, 92 comments in 6 months
Contributions summary:Siva primarily contributed to the `nnvm` project by addressing issues related to model importing, shape inference, and operator implementation within the context of deep learning. They fixed bugs in existing operators like `squeeze` and added new operators like `sum` for ONNX compatibility. Their changes also included layout conversions and optimization, showcasing an understanding of computational graph optimization.
cudametalcomputation-graphtvmdeep-learning
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.