Felix Yu
Founder at TopPick Analytics Limited
Hong Kong, China
Summary
👤
Senior🎓
Top SchoolFelix Yu is a founder and independent AI researcher with nine years of experience building data-driven fintech products and machine learning systems from Hong Kong. He founded TopPick Analytics, creator of DisclosureTracker—the market-leading information tracker for listed company disclosures—and licensed analytics tech to major Hong Kong brokers. Felix combines hands-on ML engineering (including CNN fine-tuning in Keras and contributions to architectures like ResNet and Inception) with product-led entrepreneurship, having previously co-founded startups in legal tech and consumer mobile. His background spans quantitative trading tool development at Credit Suisse, knowledge-graph quality work at Diffbot, and advanced studies in computational and mathematical engineering at Stanford. A Kaggle Master, he uniquely blends competitive modeling expertise with production-grade analytics and data curation. Colleagues describe him as a builder who moves quickly from prototype to scalable deployment while keeping a researcher’s curiosity for novel model improvements.
9 years of coding experience
3 years of employment as a software developer
B.Sc. (Advanced Track) Operations Research Concentration: Financial Engineering Minors: Applied Mathematics Economics, B.Sc. (Advanced Track) Operations Research Concentration: Financial Engineering Minors: Applied Mathematics Economics at Columbia EngineeringColumbia UniversityMaster of Science (M.Sc.) Computational and Mathematical Engineering, Master of Science (M.Sc.) Computational and Mathematical Engineering at Stanford University
English, Chinese, Chinese
Github Skills (10)
mask-rcnn10
model-building10
keras10
computer-vision10
faster-rcnn10
machine-learning10
deep-learning10
tensorflow10
python10
fine-tuning10
Programming languages (3)
Github contributions (5)
flyyufelix/cnn_finetune
Oct 2016 - Sep 2017
Fine-tune CNN in Keras
Role in this project:
ML Engineer
ML EngineerContributions:13 commits, 11 pushes, 1 branch in 10 months
Contributions summary:Felix primarily contributed to implementing and fine-tuning various Convolutional Neural Network (CNN) models within the Keras framework. Their work included adding and modifying model architectures, such as GoogLeNet, ResNet, Inception, and DenseNet, which involved significant changes to code and dependencies. Further contributions involved integrating a working example using Cifar10 for fine-tuning and addressing issues related to updating Keras/Tensorflow versions and documentation.
imagenetimage-recognitionefficientnetdeep-learningfine
flyyufelix/sonic_contest
Jun 2018 - Aug 2018
Contributions:4 commits, 2 pushes, 1 branch in 2 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.