I'm an Agentic AI Engineer with a strong foundation in Python, specializing in building intelligent multi-agent systems and workflow automation. I work hands-on with OpenAI Agents SDK, CrewAI, LangChain, and n8n to orchestrate agent communication, tool calling, and decision-making in dynamic workflows.
I engineer AI systems capable of handoffs, context sharing, and multi-step reasoning across multiple agents — and I bridge that with solid backend engineering using FastAPI (scalable APIs) and Pydantic (schema validation), so AI behavior sits on stable, production-ready infrastructure.
On the frontend/full-stack side, I build with TypeScript / Node.js, so I can ship complete agentic products end-to-end — not just the AI layer.
Currently focused on: researching agent orchestration patterns, optimizing task delegation between agents, and integrating automation pipelines into production environments.
Agentic AI & Orchestration
Backend
Tools
| Project | What it does | Stack |
|---|---|---|
| 🤖 Caring-Tutor | An AI tutor agent that teaches and looks after student wellbeing — adaptive, empathetic tutoring instead of plain Q&A. | Next.js · TypeScript · Tailwind |
| 🗣️ english-tutor-agent | A conversational AI agent that helps users practice and improve spoken/written English in real time. | Python · Agent SDK |
| 📞 customer-service-agent-for-ptcl | A CrewAI-powered support agent that classifies customer issues (billing, connectivity, router faults) and routes/responds automatically. | Python · CrewAI |
| 🤖 physical-ai-humanoid-robotics | A 12-chapter, 42,000+ word open-source textbook on Physical AI & Humanoid Robotics — ROS 2, Gazebo, Unity, NVIDIA Isaac, and VLA models, with 52 hands-on exercises. | Docusaurus · TypeScript |
