Sushanth Rajasankar
Principal Software Engineering Manager at Microsoft
Redmond, Washington, United States
Summary
🤩
Rockstar🎓
Top SchoolSushanth Rajasankar is a Principal Software Engineering Manager at Microsoft with seven years in senior engineering and leadership roles focused on web and device platforms. He leads the Web ML team for Microsoft Edge, advancing on-device ML APIs and capabilities to empower web developers. His hands-on background includes optimizing ML inference for ONNX Runtime on WebGPU—delivering measurable performance gains through specialized quantized matmul kernels. Prior experience spans Edge, HoloLens analog R&D, Surface Hub, and the Internet Explorer web platform, giving him deep expertise at the intersection of browsers, hardware, and ML. Based in Redmond, he combines product-focused leadership with low-level performance engineering and a master’s degree in computer science.
7 years of coding experience
11 years of employment as a software developer
Master's degree, Computer Science, Master's degree, Computer Science at University of Minnesota
Hindi, Tamil, English
Github Skills (8)
matrix-multiplication10
machine-learning10
onnx10
webgpu10
hardware-acceleration10
neural-network9
deep-learning9
compiler9
Programming languages (7)
Github contributions (5)
microsoft/onnxruntime
Nov 2024 - Apr 2025
ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator
Role in this project:
ML Engineer
ML EngineerContributions:39 reviews, 18 PRs, 57 pushes in 4 months
Contributions summary:Sushanth primarily focuses on optimizing and implementing machine learning inference for the ONNX Runtime, specifically targeting WebGPU. Their contributions involve developing and refining specialized kernels for quantized matrix multiplication (matmul) operations, including DP4A and SubGroupMatrix implementations. These efforts aim to improve performance, particularly for prefill and attention mechanisms, evidenced by performance benchmarks. The user's work significantly improves inference speed, as demonstrated by performance gains reported in the commit messages.
runtimetrainingtensorflowai-frameworkaccelerator
sushraja-msft/onnxruntime
Nov 2024 - Feb 2025
ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator
Contributions:26 pushes, 7 branches in 2 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.