About

I grew up in Nepal, taking things apart to see how they worked. That curiosity turned into computer engineering, and then into shipping software at startups where the gap between an idea and a working product is measured in days.

Today I work across the stack as a product engineer at DevDash Labs. I'm a core frontend and backend developer on Alkemy OS, building the Next.js application and the FastAPI services behind it — billing and credits, a unified signals feed, global search, and the onboarding and interface work that keeps dense, research-heavy workflows legible.

How I work

• Drive the AI, don't defer to it.Models write code fast, but keeping it correct is the engineer's job. I review everything that ships — velocity means nothing without verification.
• UI/UX is where the product lives.A feature nobody can find isn't a feature. I reach for interface patterns people already know rather than inventing ones they'd have to learn, and I treat responsive polish, clear feedback, and fast customer support as part of the work — not extras.
• Evals and tests are first-class. Especially with AI in the loop, evals and a real test suite are how you prove something works. I budget real time for them.
• Type-safety, always.The backend (Python, Pydantic) and the frontend (TypeScript) share the same shapes. If it compiles, it should run. That's just how I build.

Tools I reach for

• Frontend: Next.js, React, TypeScript, Tailwind, TanStack Query
• Backend and AI: FastAPI, Python, LangGraph, Pydantic, PostgreSQL
• Infra: Vercel, AWS, Docker

Away from the editor, I'm usually reading about agentic systems and how AI can do more of the tedious work so people can do more of the interesting kind.