Jiancheng Li
Senior Algorithm Engineer at Alibaba Group
Chaoyang District, Beijing, China
Summary
🤩
Rockstar🎓
Top SchoolJiancheng Li is a senior algorithm engineer with 12 years of experience building and optimizing production-ready AI systems across recommender systems and computer vision. Currently at Alibaba, he designs match-and-rank algorithms for large-scale mobile information flows, and previously contributed to model compression and NAS work at Baidu, including as a main contributor to PaddleSlim. His open-source work spans practical deep-learning tooling—refactoring loss functions in the popular mmfashion repo and implementing multi-point NMS and polygon operators in Paddle-Lite—highlighting a focus on model accuracy and inference efficiency. With internships at top labs (SenseTime, Microsoft Research Asia, DiDi, Tencent) and advanced degrees from Tsinghua and UTS, he blends rigorous research experience with hands-on system engineering. Notably, he often bridges algorithmic innovation and deployment constraints, optimizing models for edge and mobile inference in real-world products.
12 years of coding experience
2 years of employment as a software developer
Bachelor of Engineering (B.E.), Computer Science, Bachelor of Engineering (B.E.), Computer Science at National Sun Yat-Sen UniversityBachelor of Engineering - BE, Software Engineering, Bachelor of Engineering - BE, Software Engineering at Beihang UniversityMaster of Science - MS, Computer Science and Technology, Master of Science - MS, Computer Science and Technology at Tsinghua UniversityMaster of Science - MS, Computer Science, Master of Science - MS, Computer Science at University of Technology Sydney
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (11)
neural-network10
computer-vision10
pytorch10
machine-learning10
loss-functions10
deep-learning10
python10
cprogramming-language9
c-language9
paddlepaddle9
embedded8
Programming languages (2)
Github contributions (5)
open-mmlab/mmfashion
Jan 2019 - Dec 2020
Open-source toolbox for visual fashion analysis based on PyTorch
Role in this project:
ML Engineer
ML EngineerContributions:54 commits, 4 PRs, 34 pushes in 1 year 11 months
Contributions summary:Jiancheng primarily focused on refactoring and updating the core loss functions within the mmfashion project. Their contributions included refactoring the loss functions, introducing a WeightedBCELoss class, and modifying the train_AttrNet.py file to utilize the new loss function implementations. The user also updated the setup.py file to include dependencies for the project.
pytorchfashion-ailandmark-detectiondeep-learningtoolbox
PaddlePaddle/Paddle-Lite
Oct 2018 - Oct 2018
PaddlePaddle High Performance Deep Learning Inference Engine for Mobile and Edge (飞桨高性能深度学习端侧推理引擎）
Role in this project:ML Engineer
ML EngineerContributions:15 commits, 6 PRs, 6 issues in 6 days
Contributions summary:Jiancheng's commits primarily involve adding and modifying code related to the `gpc` library, specifically focusing on the implementation of "multi-point NMS" (Non-Maximum Suppression) functionality. Further commits involve implementing and testing new operators related to polygon transformations, reshape operations, and transpose operations within the Paddle-Lite framework. This work suggests a focus on model optimization and functionality within the deep learning inference engine.
inference-enginemobilebaidutensorflowfpga
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.