About me
He is a dedicated developer with a focus on machine learning, specifically in the area of recommender systems.
His expertise in this field has led him to secure internships at Tencent, where he implemented cutting-edge frameworks.
In addition to his technical prowess, Menki is a person of many passions. He finds joy in exploring new places through travel adventures ✈️, expanding his knowledge through avid reading 📚, and capturing moments through his lens as a passionate photographer 📸.
He is always willing to try new things, and keeping to learn from them.
Now, he is looking for job as a machine learning engineer.
Skill
Machine Learning : Proficient in Python language and Pytorch framework, able to design and develop stat-of-the art machine learning model .
Solidity : Familiar in Solidity language and completed the WTF Academy Solidity Advanced Course.
Other Skills: Proficient in using Linux common commands, Docker container management, Git code collaboration development. Communication with English, Cantonese, Chinese is OK. I am learning Japanese at the same time. Currently, I have received my JLPT N2 certificate.
Coding Languages:
Frameworks and Tools:
Interested:
My Selected Project
Contrastive Self-supervised Learning for Human Activity Recognition
PyTorchMar 2022 - Oct 2022
- Designed and developed a contrastive self-supervised learning framework to improve human activity recognition using IMU sensor data under low labeling rate conditions.
- Implemented data augmentation techniques and conducted extensive experiments, achieving a 19.2% average performance improvement over state-of-the-art methods.
- Contributed to the research community by submitting a paper to IEEE Internet of Things Journal and releasing the code on .
Work Experience
Machine Learning Engineer (Intern)Jun 2021 - Oct 2021
Tencent Inc.Shenzhen, Guangdong, China
- Deep Learning Research: Conducted research for the recommender system, improving the performance of the recommendation feed on the QQ browser.
- Model Design: Designed a large scare machine learning model (graph-based approach) for personalized cross-domain item embedding, enhancing the hit rate and efficiency of short video recommendations.
- End-to-end Model Development: Developed end-to-end flow projects and data pre-processing pipelines for efficient data transformation and predictive modeling.
- Evaluation Analysis: Led experiments to evaluate the impact of the item embeddings and practical scenario performance.
Publication
[PerCom'20] SilentSign: Device-free Handwritten Signature Verification through Acoustic Sensing
Mengqi Chen, Jiawei Lin, Yongpan Zou, Rukhsana Ruby, Kaishun Wu
in Proceedings of the 18th IEEE PerCom, pp.1-9, Austin, Texas, USA, March 2020. (CCF-B, AcceptedRate: ~16%)
[Paper]
Education
Shenzhen UniversitySept 2017 - Dec 2023 (Expected)
Ph.D. in Parallel Information ProcessingShenzhen, Guangdong, China
University of TorontoSept 2021 - Dec 2021
Summer school studentToronto, Canada
Shenzhen UniversitySept 2015 - July 2017
MS in Software EngineerShenzhen, Guangdong, China