A Non-homogeneous Navier-Stokes Framework for Global Logistics ETA

End-to-end machine learning project that predicts food delivery time using Gradient Boosting regression. Includes data analysis, feature engineering (distance calculation), model evaluation, and a Streamlit-based web interface.

DBSCAN-based route discovery and LightGBM ETA prediction models for informal transit. Achieves F1≥0.85 and MAE≤3 minutes.

Different types of ML projects

AI-powered Surge Pricing & ETA Optimization for ride-hailing platforms. Using demand forecasting and real-time ETA predictions, it optimizes fares, reduces wait times, and improves driver/passenger experience. Optimized for Tehran, it helps increase revenue and reduce pricing volatility.

Segmented ETA prediction model for on-demand food delivery. 5-stage pipeline (T1/T2a/T2b/T3/T4) with hierarchical Bayesian smoothing, OSRM calibration, and recency-weighted medians. Validated on 137K orders.

Production-grade ML system for food delivery ETA prediction using FastAPI, DVC, and MLflow.

PyTorch LSTM notebooks for last-mile delivery ETA prediction.

A full-stack machine learning architecture for food delivery ETA prediction, leveraging a DVC-driven pipeline, automated CI/CD workflows, cloud artifact management, and LGBM-based stacked regression ensemble for high-fidelity time estimations.

Real-Time Delivery ETA Prediction and Delay Risk Analytics

A web-based vessel tracking and maritime intelligence system using Datalastic AIS APIs for real-time vessel tracking, port monitoring, and analytics.