I build AI systems people can actually use
I'm Muhammad Adnan Tariq, a Generative AI Engineer based in Sahiwal, Pakistan. I build intelligent systems with Large Language Models, AI agents and retrieval, and I care most about the part after the demo: making them reliable enough to use every day.
I graduated in 2026 with a BS in Computer Science from COMSATS University Islamabad, Sahiwal Campus, with a CGPA of 3.80/4.00. I work in Python, FastAPI and LangChain on the AI side, and TypeScript for the products around it.
From an AI idea to a tool that holds up in real use
I combine AI engineering, product thinking and implementation discipline to build systems that are not just impressive demos, but useful tools. That means prompts and guardrails designed around real tasks, retrieval that people can trust, and interfaces that make a model's output easy to act on.
Production-minded
Guardrails, evaluation and sensible fallbacks come first, so a system keeps working when real input gets messy.
Research-backed
I write research papers on federated deep learning and interpretable medical AI, and bring the same rigour to product decisions.
Healthcare focus
Much of my recent work is in healthcare AI: medical document understanding, oncology patient support and blood-cell classification.
Experience and education
What I can help build
Four kinds of work I take on, from a first prototype to a system running in production.
LLM Applications
Chat interfaces, copilots and AI assistants designed around real user tasks, shaped for reliable production use.
- Assistants built around a specific job, not a generic chat box.
- Prompt design and guardrails for predictable output.
- Structured responses that plug into existing products.
- Interfaces that show sources and make answers easy to check.
Track record
A few numbers, each with something behind it: the papers and the hackathon build are on this site for you to read.
Where I've studied, worked, competed and published
- COMSATS University IslamabadBS Computer Science
- AI 3D Scanning SolutionsAI engineering internship
- KaggleMedGemma Impact Challenge
- Tech Science PressPublished research