projects lab
Systems built end to end, not just written about: control frameworks, forecasting pipelines, and the tooling around them. Each entry links to code or data where the work is public.
4 entries
IBM · Machine Learning & Data Science
Built propensity models on more than 65M behavioral records, supported by statistical testing, feature ablation, and model benchmarking. Integrated predictions into an LLM-powered application for natural-language audience definition and model-driven decision support.
Lightsense Technology · Machine Learning Research
Developed and benchmarked multimodal deep learning models for drug detection from spectral sensor data. Evaluated robustness under noise and distribution shift and used predictive modeling to improve sensing configurations and detection sensitivity.
DIDO · Machine Learning Engineering
Developed PPO-based reinforcement learning policies for intelligent building energy control. Worked across the full ML lifecycle, including simulation, offline policy evaluation, AWS deployment, and production monitoring.
Digital Attitude · Data Science
Built data pipelines and predictive systems for behavioral analytics, anomaly detection, and operational monitoring. Worked across ETL, feature engineering, predictive modeling, and production analytics.