Buildingintelligent systems.
GenAI & Agentic AI Developer. I design and build AI-powered applications, agentic workflows and reliable Python backends using modern Generative AI technologies.
03+
AI Projects
03+
Live deployments
12+
Technologies
Open
Internships & full-time
Selected work
Projects that solve problems.
A selection of AI and backend systems I have designed and built.
Under the hood
How my agents work.
These flows are taken from the real code of my LangGraph projects. Click any step to see what it does.
An SRE agent that diagnoses incidents and proposes fixes.
Step 1 of 5 · Incident report
The incident or alert description enters the system.
Live demo
Watch an agent think.
This is not an animation, it is a real agent. Ask a question: it finds relevant projects in the database, writes the answer with a Groq LLM, and then checks whether the links in the answer actually exist in the portfolio. Every step is streamed live from the backend.
Question
Input
Retrieve
Keyword search · DB
Draft
LLM · Groq
Verify
Link grounding check
Live trace
The agent is idle. Ask a question above or tap a suggestion.
Live status
Are the demos online right now?
This board is real: my server actually pings every demo. On free hosting, demos fall asleep after a while, and this check wakes them up too, so there is less waiting after you press "Open demo".
Under the hood
How this site works.
This portfolio is a small production system in itself. Pick a flow below and watch how a request travels through it.
Step 1 / 6
Your question goes straight to FastAPI.
Hands-on
Be the on-call engineer.
This is a simulation of my Incident_AI agent. Pick an incident, watch how the agents diagnose it, and approve or reject the fix yourself, just like the human-in-the-loop gate in the real agent.
Ready. Pick an incident and trigger it.
- Diagnose
- Propose fix
- Approval
- Execute
Incident report
SEV-102:14 UTC: checkout-service error rate jumped from 0.2% to 38% right after the 02:05 deploy (v2.31). Logs show NullPointerException in PaymentMapper. Database and queue metrics look healthy.
ISH Digital Studio
I also build websites.
Concept websites for local businesses, built under my freelance studio. These are demo builds I make to pitch to businesses, not paid client projects.
Technology
Tools I build with.
A practical stack focused on AI engineering, backend development, APIs, databases and deployment.
About me
Building with curiosity.
My focus is on turning AI capabilities into useful applications with clean APIs, reliable backend architecture and intelligent agent workflows.
Open to remote & on-site opportunities
Have an ideaworth building?
I build AI agents, RAG systems and FastAPI backends. If you are hiring, or have a problem that needs an intelligent system, send me a note. I read every message.