N K
Lead Software Engineer - MONAKA Software R&D Unit (FRIPL) at Fujitsu Research
Bengaluru, Karnataka, India
Summary
👤
Senior🎓
Top SchoolN K is a Lead Software Engineer specializing in AI frameworks and performance optimizations, with nine years of experience tuning PyTorch, ONNX Runtime, OpenVINO and JAX for x86 and ARM platforms. Based in Bengaluru, he leads Fujitsu Research's MONAKA R&D unit, driving advances in quantization, sparsity, multi-threaded scaling and ScientificML for HPC and next‑gen supercomputing. He has a strong open-source footprint—notably contributions to Microsoft’s ONNX Runtime and example repos, including improving the OpenVINO Execution Provider for multi-threading, hetero support and fp16 performance. Before Fujitsu he built integrations and end-to-end AI workflows at Intel, earned a patent for ensemble-based cluster tuning, and helped enable SVE support in PyTorch to accelerate millions of ARM devices. Colleagues describe him as a hands-on tech lead who mentors engineers while shipping low-level optimizations that unlock real-world inference and training gains. He combines systems-level C++/Python expertise with practical deployment experience across cloud, edge and VPU hardware.
8 years of coding experience
6 years of employment as a software developer
Master's degree Software Engineering, Master's degree Software Engineering at Vellore Institute of Technology
English, Hindi, Telugu, Urdu
Github Skills (15)
computer-vision10
machine-learning10
deeplearning-ai10
c-language10
deep-learning10
onnx10
cprogramming-language10
onnxruntime10
python10
hardware-acceleration10
openvino10
yolov49
object-detection9
tensorflow5
pytorch5
Programming languages (3)
Github contributions (5)
microsoft/onnxruntime-inference-examples
Oct 2021 - Apr 2022
Examples for using ONNX Runtime for machine learning inferencing.
Role in this project:
ML Engineer
ML EngineerContributions:4 reviews, 5 commits, 2 PRs in 6 months
Contributions summary:N primarily contributed to the development and maintenance of example code for object detection using the ONNX Runtime with the OpenVINO Execution Provider. Their work included updating sample code, specifically for YOLOv4, to align with changes in the ONNX Runtime API. They also added new samples, corrected errors, and updated the documentation for the examples. Furthermore, they focused on ensuring the code functions correctly across different platforms, specifically including Windows.
deep-learningruntimeinferencemachine-learningonnx
microsoft/onnxruntime
Oct 2020 - Apr 2022
ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator
Role in this project:ML Engineer
ML EngineerContributions:60 reviews, 106 commits, 36 PRs in 1 year 5 months
Contributions summary:N primarily contributed to the OpenVINO Execution Provider (EP) for ONNX Runtime. Their work focused on enabling and improving multi-threading support within the OpenVINO EP, specifically for the OpenVINO EP and VPU hardware. They also implemented Hetero support, added security checks, fixed batching logic, and enabled OpenVINO Runtime options for the perftest application. Furthermore, they addressed bugs and improved the performance of various models within the OpenVINO EP, including enabling fp16 for input types.
runtimetrainingtensorflowai-frameworkaccelerator
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.