You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Smart Grid AI System: A real-time energy monitoring and forecasting dashboard. Features live electricity demand tracking (EIA API), machine learning load prediction using Random Forest, and grid anomaly detection via Isolation Forest. Built with Flask, scikit-learn, and Chart.js.
Fault detection on offshore well sensor data. Python pipeline ingests production and telemetry records, flags anomalous wells, and surfaces results in a Streamlit dashboard.
End-to-end credit card fraud detection pipeline with preprocessing, feature scaling, and class imbalance handling via undersampling and SMOTE. Trains Logistic Regression and Decision Tree models, evaluates with key metrics, and exports the final model using joblib for manual transaction classification.
Mahalanobis scoring stage of PaDiM anomaly detection, implemented four ways - PyTorch, C++17 CPU, hand-written CUDA (naive + PBCR), cuBLAS - behind a fixed artifact boundary, with a 1e-4 numerical agreement gate on every reported timing.
Explainable anomaly detection for NOAA-GLERL Lake Erie buoy data (2015–2026): dead-probe, spike, stuck and neighbour-buoy checks in Pandas and PySpark, compared with operator quality flags. Poster for LAKE 2026.
Structured machine learning notes, with mathematical explanations and visual breakdowns of core algorithms - from regression and neural networks to clustering, recommender systems, and reinforcement learning.