Projects

Portfolio-grade experiments with real constraints.

Traffic Flow Forecasting

Forecasts short-term roadway demand via a lightweight ML pipeline that can run on commodity hardware.

Pairs CSP reasoning with heuristic pruning so every move can be traced and explained on randomized boards.

Explores ranking strategies via deterministic graphs, sampling approximations, and convergence diagnostics.

Designed and trained a CNN to classify traffic signs from raw images with reliable generalization on unseen data.

Physics-informed simulator generating synthetic STEM nanocluster images with deterministic reproducibility for machine learning research.