Summary
Zimeng Lyu is an Assistant Professor and machine learning researcher with nine years of experience specializing in time series forecasting, continual learning, neural architecture search, and stock return prediction. He designs cross-domain forecasting models for edge devices and has translated academic work into real-world impact—his evolved recurrent networks helped prevent roughly $7.3M in revenue losses at power plants and his portfolio strategies have beaten the market. Zimeng developed ONE-NAS for real-time forecasting and the EXAMM NeuroEvolution algorithms, blending online learning with neuroevolution techniques that enable adaptive, production-ready models. He teaches AI and advanced software engineering, builds practical ETL/DevOps workflows for data science, and holds a PhD from RIT with a strong record of applied ML research and a patent-linked background in aerial image analysis.
9 years of coding experience
7 years of employment as a software developer
Doctor's Degree, Time Series Forecasting, Online Learning, Stock Return Prediction, 3.92, Doctor's Degree, Time Series Forecasting, Online Learning, Stock Return Prediction, 3.92 at Rochester Institute of Technology
Bachelor's degree, Electrical Engineering, Bachelor's degree, Electrical Engineering at Nanjing Agricultural University
Master's degree, Computer Engineering, 3.7/4.0, Master's degree, Computer Engineering, 3.7/4.0 at Syracuse University
English, Chinese