[Project one]
[2026][One or two lines: the problem and what you built.]
I build products end to end, and the AI inside them.
Role
Full-stack · AI engineer
Experience
2 years in production
Local time
--:-- IST
Now playing
[Track] — [Artist]
[One or two lines: the problem and what you built.]
[One or two lines: the problem and what you built.]
Software Developer · AI
[What you own, the AI features you shipped, and one result.]
Software Developer
[What you built full-stack, and the impact.]
[Internship / freelance / education]
[One line.]
[1st]
[Hackathon name]
[What you built in 36 hours, in one line]
[2025]
[Top 5]
[Hackathon name]
[What you built in 36 hours, in one line]
[2025]
[Finalist]
[Hackathon name]
[What you built in 36 hours, in one line]
[2024]
"[A real line from a manager about how you work.]"
"[A real line from a teammate or client.]"
Build
[Front-end tool] [Back-end tool] [Database]
AI
LLM applications Retrieval (RAG) Agents & evals
Design
[Design tool] Prototyping Design systems
Ship
[Cloud] [CI/CD] [Monitoring]
Clarity over cleverness
If a user needs a manual, the design isn't finished. I start from the problem, not the stack.
AI with a reason
I add AI where it saves real time or removes real friction, and I measure that it does.
Finish the last 10%
Edge cases, empty states, the bug at 1 am. That last stretch is where quality lives.
I'm a full-stack developer who moved into AI engineering. I like software that feels calm to use and holds up in production, and I don't leave things half done.
That thin line on the left of this page? It's one long thought. Scroll to the end to see how it finishes.