Kian Gan
System Software Development Engineer at Intel Corporation
Kuala Lumpur, Malaysia
Summary
🤩
Rockstar🎓
Top SchoolKian Gan is a system software development engineer based in Kuala Lumpur with two years of professional experience building low-level software stacks for vision and neural processing units at Intel. He integrates deep learning frameworks and compiler/device plugins, and has improved backend infrastructure for the notable open-source OpenVINO toolkit by enhancing logging, fixing memory and model retrieval issues, and addressing static-analysis findings. Kian has also built tooling that speeds computation—writing a JOCL-based Java program to offload CPU work to accelerators—and delivered web apps for test plan and performance management. His background includes microservices and mobile work, reflecting a practical full-stack mindset applied to hardware-near software problems. A first-class Computer Science graduate from UTAR, he combines attention to code quality with a focus on performance and observability in production AI inference stacks. Notably, his contributions emphasize diagnosability and memory optimization—small changes that materially improve stability for NPU deployments.
2 years of coding experience
Bachelor of Computer Science, Bachelor of Computer Science, Honours Degree with Distinction (First Class Honours), Bachelor of Computer Science, Bachelor of Computer Science, Honours Degree with Distinction (First Class Honours) at Universiti Tunku Abdul Rahman (UTAR)
Github Skills (12)
debugging10
logging10
debug10
c-language10
backend10
cprogramming-language10
back-end-development10
openvino10
performance-optimization9
ai9
deep-learning8
computer-vision8
Programming languages (1)
Github contributions (3)
openvinotoolkit/openvino
Nov 2023 - Nov 2024
OpenVINO™ is an open source toolkit for optimizing and deploying AI inference
Role in this project:
Backend Engineer
Backend EngineerContributions:25 reviews, 11 PRs, 1 branch in 1 year
Contributions summary:Kian primarily contributed to the backend infrastructure of the OpenVINO toolkit, focusing on the Intel NPU plugin. Their work involved adding logging and debugging messages to improve error verbosity and diagnose issues within the `zero_backend` components. They fixed model name retrieval and addressed a Coverity issue, demonstrating a focus on code quality and stability. Furthermore, they optimized memory usage by freeing serialized models earlier in the compilation process.
inference-enginepytorchmodel-optimizerdeep-learninggpu
KianYongGanIntel/openvino
Nov 2023 - Nov 2024
OpenVINO™ is an open-source toolkit for optimizing and deploying AI inference
Contributions:124 pushes, 23 branches in 1 year
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.