Yaowen Liu
Machine Learning Engineer at EvenUp
Old Toronto, Ontario, Canada
Summary
👤
Senior🎓
Top SchoolYaowen Liu is a Machine Learning Engineer with 8 years of experience and 5+ years focused on building and deploying production AI systems across infrastructure, vector search, and LLM/VLM applications. He combines strong software engineering in C++, Python and SQL with hands-on deployment skills—shipping high-throughput services (1,000+ QPS) and scalable feature inference using TensorRT and gRPC. Yaowen has led end-to-end projects from prototype to wide deployment, including a person re-identification system used across 10,000+ stores and on-premise GenAI assistants that improved call center efficiency. He’s delivered measurable gains (e.g., 55% deployment cost reduction via feature quantization, 97% mAP on cross-camera matching) and built monitoring and alerting to sustain 99% data delivery SLAs. A UofT MEng with practical research-to-production experience, he also tunes LLMs (LoRA) and VLM pipelines to boost accuracy and throughput, reflecting a rare blend of model engineering and production optimization. Yaowen is driven to own products end-to-end, favoring pragmatic, measurable solutions over academic novelty.
8 years of coding experience
5 years of employment as a software developer
Bachelor's degree, Earth Information and Technology, Bachelor's degree, Earth Information and Technology at Sun Yat-sen UniversityMaster of Engineering - MEng, Mechanical & Industrial Engineering - Data Analytics and Machine Learning, Master of Engineering - MEng, Mechanical & Industrial Engineering - Data Analytics and Machine Learning at University of Toronto
Github Skills (72)
cross-domain10
image-retrieval9
metric-learning9
computer-vision8
resnet8
xterm8
python8
deep-learning8
gpu8
zmodem8
machine-learning8
re-identification8
telnet7
windows7
bash7
Programming languages (4)
Github contributions (5)
TheBobbyliu/Person-ReID
Mar 2019 - Apr 2020
Pytorch Person ReID, including implementation of MGN, PCB, RPP, ResNet, MobileNetV2, and so on.
Contributions:18 commits, 31 pushes, 1 branch in 1 year 1 month
pytorchperson-reidpersonpcbmobilenetv2
TheBobbyliu/Dog_Prediction_with_CNN
Aug 2018 - Apr 2019
A dog prediction project using CNN, an initial accuracy is around 84%.
Contributions:2 pushes, 1 branch in 7 months
limepythonpredictiondeep-learningmachine-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.