Xuesong Yang
United States
Summary
🤩
Rockstar🎓
Top SchoolXuesong Yang is a quantitative researcher and masters candidate in Quantitative and Computational Finance at Georgia Tech with eight years of experience building multi-factor strategies across equities and futures. He has hands-on expertise in factor mining, enhancement, and evaluation—having built a personal factor library spanning high-frequency to low-frequency and alternative factors—and applies machine learning and deep learning for time-series forecasting and signal enhancement. In industry roles he improved model performance (e.g., a 20% uplift vs. baseline LightGBM), sped up genetic programming by 60%, and achieved strong backtested returns for futures strategies. He is proficient in Python and C++, comfortable with large-scale cross-sectional and HFT data, and has published engineering contributions to few-shot learning tooling on GitHub (LibFewShot). Less obvious: he has applied generative techniques (TimeGAN / counterfactual data) to mitigate overfitting and even co-filed patents on counterfactual sample generation and data quality evaluation. Seeking a summer internship, he blends rigorous research, production-minded engineering, and practical trading experience.
8 years of coding experience
1 year of employment as a software developer
学士, 计算机与金融工程, 学士, 计算机与金融工程 at 南京大学Master's degree, Qutitative and Computational Finance, First garde, Master's degree, Qutitative and Computational Finance, First garde at Georgia Institute of Technology
Github Skills (5)
pytorch10
image-classification9
python9
mask-rcnn9
faster-rcnn9
Programming languages (3)
Github contributions (5)
RL-VIG/LibFewShot
Dec 2020 - Nov 2022
LibFewShot: A Comprehensive Library for Few-shot Learning. TPAMI 2023.
Role in this project:
ML Engineer
ML EngineerContributions:175 commits, 4 PRs, 55 pushes in 1 year 11 months
Contributions summary:Xuesong primarily contributed to the model and training code within the LibFewShot repository. Their commits focused on resolving issues related to target shape mismatches and fixing meta-learning components, especially in the ANIL model. Additionally, the user added new code for DSN (Distance Similarity Network) and made updates to finetuning models like Baseline, BaselinePlus, MTLPretrain, and SKDModel, indicating involvement in various few-shot learning methods and model implementations.
pytorchmeta-learningdeep-learningshot-learningcomputer-vision
yangcedrus/QingXie
Oct 2017 - Apr 2021
Formal
Contributions:11 PRs, 32 pushes, 2 branches in 3 years 6 months
formal
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.