You matched with
Guoqing Hu
Guoqing Hu
Institution
Cornell University
Introduction
Guoqing enjoys working at the intersection of technology and real-world problem solving, especially in areas like machine learning, optimization, and data-driven systems. Outside of work, he spends a lot of time with his family, especially his dogs and cats, and enjoys being outdoors, exploring trails, spending time by the lake, and learning new things. He describes himself as easygoing, curious, and interested in both deep technical work and simple, fun moments in everyday life.
Top Fields
Interdisciplinary / Emerging Fields, Computer Science, Engineering
Research Areas
This mentor is comfortable supporting projects involving machine learning model training, building machine learning systems for real-world applications, and multi-dynamic models. Their interests sit at the intersection of machine learning, aerodynamics, and mathematics.
Background
Guoqing’s research background includes formulating and solving nonlinear optimization problems for complex dynamic systems using model predictive control and gradient-based solvers such as CVXPy and IPOPT. He has integrated physics-informed models into optimization frameworks, benchmarked solver performance across linear, quadratic, and nonlinear programming formulations, implemented adaptive constraint strategies, and developed simulation frameworks to validate optimization quality against experimental and synthetic datasets.
He has also worked extensively with large-scale weather and operational data, including collecting, processing, and analyzing millions of records from forecasts, historical logs, and sensor data. His experience includes building reproducible data workflows in SQL and Python, transforming unstructured and semi-structured data into structured features, and using feature engineering and data validation to support model calibration, optimization, and performance analysis.
• Formulated and solved nonlinear optimization problems for complex dynamic systems using
model predictive control (MPC) and gradient-based solvers (CVXPy, IPOPT).
• Integrated physics-informed models into optimization frameworks to enhance accuracy and
convergence stability.
• Benchmarked solver performance across different formulations (LP, QP, NLP) and implemented
adaptive constraint strategies to improve computational efficiency.
• Developed and validated simulation frameworks to assess optimization quality against high-fidelity
experimental and synthetic datasets.
• Collected, processed, and analyzed millions of weather records from heterogeneous sources
(forecasts, historical logs, sensor data), transforming largely unstructured and semi-structured
data into structured features for control and optimization models.
• Built reproducible data workflows using SQL and Python to query, clean, and aggregate large-
scale weather/operational data for model calibration, validation, and performance analysis.
• Performed feature engineering and data validation on high-dimensional clima