Summary
Siming Liu is a data scientist and PhD candidate in Geophysical Fluid Dynamics with 11 years of experience applying machine learning, statistical analysis, and scalable cloud workflows to climate science, venture-capital analytics, and environmental risk. She builds production-ready ML systems—ranging from NLP pipelines that predict early-stage startup success to hybrid physics–data models that deliver ~1000× faster extreme-weather forecasts than traditional dynamical models. Her work blends rigorous statistical testing (e.g., McNemar’s tests on AI weather models) with large-scale data engineering (processing TB-scale climate simulations on GCP/BigQuery) to improve model interpretability and operational impact. Comfortable bridging research and product, she has deployed mobile health vision models in production and led a $65K summer school, demonstrating strong program management and cross-functional leadership. Fluent in Python, PyTorch/TensorFlow, XGBoost, and climate toolkits (Xarray, CMIP6, ERA5), she seeks roles that turn complex geophysical data into actionable decisions. An uncommon strength is her track record of integrating legacy Fortran physics codes with modern gradient-based optimization to squeeze both speed and fidelity from forecasting systems.
11 years of coding experience
3 years of employment as a software developer
Bachelor of Science - BS, Atmospheric Science, 91.3/100, 4.18/5, Bachelor of Science - BS, Atmospheric Science, 91.3/100, 4.18/5 at Nanjing University of Information Science and Technology
Jinling High School
PhD Candidate, Geophysical Fluid Dynamics, PhD Candidate, Geophysical Fluid Dynamics at University of Chicago