Muhammad Abdullah

Agentic AI Engineer · Cloud Native Developer

---

I build multi-agent AI systems and cloud-native backends that run in production.

Most of my work sits at the intersection of autonomous agent orchestration (LangGraph, CrewAI), retrieval-augmented generation, and infrastructure (FastAPI, Docker, AWS, Azure). I've shipped agent pipelines with up to 18 coordinated agents serving real users, and I deploy them behind Traefik with subdomain-based routing on managed cloud infrastructure.

I also teach what I build. I run The College of AI on YouTube and have taught Python, Agentic AI, Docker, and N8N to 500+ students across multiple cohorts.

---

Currently working on

• Multi-agent orchestration systems using LangGraph with complex state management and human-in-the-loop workflows
• Cloud-native service deployments with FastAPI, Docker, Traefik, and AWS (EC2, RDS, Route53, ACM)
• RAG pipelines with custom chunking strategies across FAISS, Qdrant, and ChromaDB
• Workflow automation with N8N for content repurposing and business process orchestration

---

Tech Stack

---

Selected Projects

Project	Description	Stack
Observator	Government-grade B2G labour market intelligence platform with an 18-agent LangGraph pipeline, serving real-time workforce analytics with UAE data sovereignty compliance	React 18, FastAPI, PostgreSQL, DuckDB, TimescaleDB, Qdrant, Azure AKS
NextMovePath	Multi-agent immigration consultant system with eligibility assessment, CRS scoring, document checklist generation, and RAG over official IRCC guidelines	LangGraph, FastAPI, FAISS, GPT-4
DocumedAI	AI-powered medical documentation platform that automates clinical note generation and report interpretation using multi-modal RAG	FastAPI, LangChain, OpenAI, RAG

---

Explore my work

The best way to see what I'm building is to look at the code itself. Check out my pinned repositories below for production-grade agent systems, RAG pipelines, and cloud-native backends.