Chun-i Tsai
Staff Machine Learning Engineer at Netskope
Taoyuan City, Taiwan
Summary
🤩
Rockstar🎓
Top SchoolChun-i Tsai is a Staff Machine Learning Engineer based in Taoyuan City, Taiwan, with over five years of experience deploying AI models and optimizing edge inference on SoCs. He has driven production LLM deployments (including LLAMA2/3) at Qualcomm and contributed to the open-source compiler ecosystem—helping improve TVM’s frontend span tracing and adding a Qualcomm backend to Executorch. At Qualcomm he built TVM Explorer, a GUI for source mapping across ML frameworks, and mentored interns on compiler and runtime tooling. Now at NetSkope, he focuses on scaling ML inference for security products while bringing deep expertise in compiler engineering and embedded AI. An academic background from Uppsala University and top-grade graduate work in Taiwan underline his blend of rigorous research training and hands-on systems delivery.
5 years of coding experience
6 years of employment as a software developer
碩士, 資訊工程系研究所, 4.3/4.3, 碩士, 資訊工程系研究所, 4.3/4.3 at 國立臺灣師範大學Master's degree, Information technology, Master's degree, Information technology at 瑞典烏普薩拉大學
Github Skills (12)
compiler10
relay-modern10
tvm10
compiler-compiler10
relay10
python9
deeplearning-ai9
machine-learning9
deep-learning9
tflite8
pytorch8
tensorflow8
Programming languages (2)
Github contributions (5)
apache/tvm
Apr 2021 - Jan 2023
Open deep learning compiler stack for cpu, gpu and specialized accelerators
Role in this project:
Back-end Developer / Compiler Engineer
Back-end Developer / Compiler EngineerContributions:26 reviews, 7 commits, 14 PRs in 1 year 9 months
Contributions summary:Chun-i's primary contribution involves fixing grammar issues in the runtime documentation. They also worked on adding a span filling feature for frontends in Relay, specifically for TensorFlow, TFLite, and PyTorch. This involved modifying Python files related to the frontend and adding tests to verify the span functionality. Their work focused on enhancing the debugging and tracing capabilities within the TVM compiler stack.
metalvulkancompilertensoropencl
chunit-quic/tvm
Sep 2021 - Mar 2023
Open deep learning compiler stack for cpu, gpu and specialized accelerators
Contributions:8 commits, 78 pushes, 23 branches in 1 year 6 months
cpugpu-programminggpu-accelerationtvmdeep-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.