Projects

An end-to-end computer vision system for aerial imagery, implemented and validated on a commercial drone for real-time video inference & counter UAS interceptor simulation

An end-to-end predictive maintenance system for aircraft turbofan engines. RUL prediction on NASA CMAPSS dataset using Random Forest, XGBoost, sklearn Pipelines & MLflow.

Sketch-to-Face Translation, concepts & architectures. Iterating from autoencoders to U-Net to Pix2Pix (conditional GAN).

Intelligent electric vehicle charging optimization using LSTM emission forecasting and convex optimization to minimize CO2 emissions while ensuring user mobility needs.

A comprehensive research study analyzing 2.4M+ users’ mobility patterns to inform California’s 2030 GHG reduction policy - using machine learning, geospatial big data analysis and network science.