AI Engineer & Full-Stack Developer
I build the systems between web software and applied AI.
I’m Ahmad Irfan — a full-stack web developer moving deliberately into AI engineering, currently formalizing that shift through an M.S. in AI & Autonomous Systems and the IBM AI Engineering specialization. This site is a running log of what I build along the way.
Ahmad Irfan@web:~$
STATUS
open_to_work
FOCUS
ai / ml engineering
STACK
next.js · python · pytorch
BASED
remote / islamabad
// selected work
Featured projects
A mix of AI/ML systems, full-stack products, and mobile apps — shipped, not just prototyped.
A retrieval augmented maths teacher for grade 1 - 6 students. The project uses Vector DB, Gemma 2B with a context window to provide accurate multi turn answers.
A GNN combined with MARL for coordinating drone swarms. The project is an implementation of QMIX-GNN and introduces some new features such as anomaly detection.
A hardware based AI Grader for MCQs using Gemini API for accurate grading. The project uses ESP32, TFT Screen and ESP Cam for hardware interfacing.
// stack.json
What I work with
Full-stack fundamentals paired with a growing applied ML toolkit.
AI / ML
Languages
Web
App
// changelog
From the blog
Notes from the transition — what I'm learning, building, and getting wrong along the way.
Teaching a Drone Swarm to Catch Its Own Rogue Agent
Aug 8, 2026 · 8 minNotes from a GNN project — coordinating a drone swarm with GNNs, and getting rogue-agent detection to 100% with zero false positives.
Building a Smart Math Tutor With a Small Model
Aug 3, 2026 · 6 minWhy I chose retrieval, conversation memory, and prompt engineering over simply using a larger language model.
// let’s build something
Hiring for an AI engineering role, or have a project in mind?
I’m currently open to full-time AI engineering roles and select freelance projects across web, app, and applied ML.