Industry Experience

My industry work focuses on large-scale machine learning systems, geospatial modeling, and production-ready AI experimentation.

Incoming ML Intern — Regeneron

Summer 2026

Joining Regeneron’s Machine Learning team to work on applied modeling and validation systems within large-scale biomedical and operational pipelines.

AI Associate Developer — Insurity

Oct 2025 – Jan 2026

Peril Prediction Project

  • Engineered large-scale geospatial preprocessing pipelines (6M+ rows) integrating climate variables with peril events via time alignment, spatial filtering, and feature engineering using Python, Pandas, and GeoPandas.
  • Built and evaluated LightGBM-based multi-class peril classifiers, applying SMOTE class balancing to improve accuracy and F1 score by approximately 10%.
  • Experimented with a Temporal Fusion Transformer (TFT), modeling seasonal dependencies via cyclical week encoding and training GPU-accelerated PyTorch models on NVIDIA RTX hardware.

Computer Vision — Geospatial Change Detection

  • Developed a geospatial change-detection system for before/after satellite imagery using OpenCV and deep learning pipelines.
  • Implemented CNN and Transformer-based architectures including U-Net++ for high-resolution semantic segmentation.