I've built four end-to-end AI projects across legal retrieval, medical transport dispatch, code review, and transformer training.
For PakLex v2, I modeled 1,000+ Pakistani statutes, sections, and case links in Neo4j, combining graph retrieval, ChromaDB vector search, and the Claude API to produce cited legal answers.
I designed SmartRide NEMT's AI dispatch pipeline using DistilBERT for patient-triage classification, XGBoost for ride-demand forecasting, and PostgreSQL/PostGIS for hospital matching. I also built a multi-agent PR review pipeline that uses the GitHub API to review diffs and post structured feedback.
I'm completing a BS in Artificial Intelligence at Iqra University and have also managed IT records and technical documentation during a six-month internship at Pakistan Television.

