Lisa Jiang
Business Support Manager at Intel Corporation
China
Summary
🤩
RockstarLisa Jiang is a Business Support Manager at Intel with 7 years of professional experience and a longer tenure at the company dating back to 2008, blending operational rigor with cross-functional program support. Based in China, she pairs corporate business management skills with hands-on technical contribution as a backend and performance engineer on high-profile open-source LLM infrastructure (vLLM), where she improved CPU inference backends and optimized memory-efficient KV cache formats. That unusual mix of business operations and low-level performance work lets her translate technical trade-offs into actionable business decisions and smoother engineering workflows. Colleagues rely on her for maintaining CI/build stability and for pragmatic fixes that keep complex systems performant in production. She has a track record of driving efficiency both in large enterprise settings and in open-source projects that underpin modern AI deployments.
7 years of coding experience
Github Skills (10)
cpu10
c-language10
inference10
cprogramming-language10
python10
llm10
openmp9
cicd7
docker6
dockers6
Programming languages (3)
Github contributions (5)
vllm-project/vllm
Sep 2023 - Apr 2025
A high-throughput and memory-efficient inference and serving engine for LLMs
Role in this project:
Back-end & Performance Engineer
Back-end & Performance EngineerContributions:91 reviews, 45 PRs, 199 comments in 1 year 6 months
Contributions summary:Lisa primarily focused on enhancing the CPU backend of the vLLM project, contributing to its performance and maintainability. They added a CPU inference backend, optimized the existing CPU backend, and integrated support for compressed-tensor W8A8 and FP8 KV cache. Additionally, the user made several code improvements, including fixes for the OpenMP settings, fixing compilation errors, and ensuring proper behavior in CI/Build, reflecting a focus on both functionality and efficiency.
amdcudadeepseekgpthpu
bigPYJ1151/vllm
Sep 2023 - Apr 2025
A high-throughput and memory-efficient inference and serving engine for LLMs
Contributions:525 pushes, 96 branches in 1 year 6 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.