Sharjeel Jalil Albasit
I build data products for pricing, prediction, and operational decision-making.
ETA Prediction Model
5-stage segmented ETA with hierarchical Bayesian smoothing, OSRM calibration, and recency-weighted medians. Validated on 137K orders, MAE 8.49 minutes.
Credit Risk Scoring and Limit Engine
MLP neural network scoring 891K bank customers with 5-tier risk categorization, proxy income estimation, and credit limit determination.
Demand Control System
Zone-level stress management with progressive search and listing radius shrink, Butler shutdown, and customer-facing delivery Choice tiers.
Courier Shift Optimization
Demand-aligned shift scheduling replacing fixed templates. Median demand forecasting, greedy selection, dispatch simulation. Utilization 0.8 to 1.1 orders/hour.
Vendor Churn Prevention Engine
Predicts which restaurants will go inactive using operational KPIs. Intervention on flagged vendors reduced churn by roughly 62% across 17K vendors.
Customer Pricing Intelligence
Infers customer personas from trip geography using OpenStreetMap, then predicts willingness-to-pay. Outputs pricing multipliers from 1.00x to 1.15x.
Smart Ride Assignment Engine
Ride acceptance prediction replacing nearest-driver dispatch with ML-driven ranking using driver behavioral odds, route patterns, and geometric features.
