Tell the counter who's travelling and where to. The board will put your train up, and the ticket is yours.
I grew up in a small town on the Bihar–Nepal border, where nobody I knew had internet at home. So as a kid I rooted an old Android phone, figured out how the Wi‑Fi networks around town worked, and got my friends and me online for free. I didn't know it was called hacking. I just knew I'd made something work that wasn't supposed to.
Today I'm a backend-heavy full-stack engineer in Hyderabad. I design systems, direct coding agents to build them, and own the review, testing and rollout myself.
When I picked a B.Tech in Computer Science, I had never heard the word coding and didn't own a laptop. I got both at Lovely Professional University. Competitive programming taught me to think fast under a timer.
Somewhere along the way I started teaching too: live sessions and tutorials on Code with Newton for 1,200+ students, plus beginner projects and guides on GitHub and YouTube.
Before my final year began, I landed a year-long internship at MAQ Software in Hyderabad, so I spent fourth year shipping software instead of sitting in class.
At an early-stage startup I built a TV dashboard for Android TV, a responsive feed and a custom in‑app camera. Then I benchmarked React Native UI libraries and led the migration to a faster one.
I designed REST APIs in Node.js and Express with validation, centralized error handling, JWT auth and Mongoose models. On the frontend I built a drag-and-drop form builder that 100+ users relied on.
I built the LLM pipeline inside Hippo, Zeta's Microsoft Teams incident bot. During a live incident it drafts customer updates on a schedule set by severity. An LLM validator checks each draft and can regenerate it up to 3 times, then an SRE approves it. The update is published exactly once, even with several replicas running.
I also built SaaS Navigator, a multi-tenant SRE console across production zones, frontend and backend, and wrote 80% of the frontend app's commits.
Each one is in production or close to it, with real infrastructure: auth, payments, migrations, app store releases.
An LLM rewrites a page into natural speech and explains code instead of reading out symbols, with highlighting synced to the voice. It speaks English, Hindi and Hinglish. It has a backend with Google sign-in, usage metering and a shared Postgres cache.
A private notes app with a biometric vault and AI memory that finds what you meant, not only what you typed. Built for both iOS and Android, with over-the-air updates.
A menstrual-health app that is live with real users. I built the Expo app and the Node backend, and I run its production database, releases and incident fixes.
A protocol for AI-written code: plan, build, review and prove, with the planner, builder and reviewer kept as separate agents so no agent grades its own work. It works with Claude Code, Codex, Gemini CLI and Cursor.
Small, bright mobile games such as Ice Smash, Egg Catcher and Duck Rescue. I handle ads, analytics and store releases. Ice Smash is live on the App Store.
Mountain roads in Himachal and Kashmir, islands in Thailand, and every dog that wanted to be in the photo. Scroll sideways for the view.
I work AI-native: I write the design, coding agents build and test it, and I own the review and the rollout. Every change has to pass these signals in order.
A written design with acceptance criteria, before any code.
A separate reviewer reads it. Gaps go back to the planner.
Agents implement against the plan, with tests for every requirement.
Evidence for each requirement, a final review, and a rollout I watch.
One-way ticket, valid for one hire
Thanks for riding the whole line. The next stop is your team.