# Himanshu Khairnar

> Himanshu Khairnar is a full-stack & AI engineer in Mumbai building LLM-powered products with Next.js, TypeScript and Python — AI commerce at RankUp, client AI platforms for Tracon, and agentic MCP integrations.

I build full-stack web products, mostly with Next.js, TypeScript and Python. Lately that means a lot of LLM-powered things: agents that read inboxes, chat that can check out a cart, dashboards that answer questions in plain English.

I like owning a thing end to end — talking to the people who'll use it, scoping it, building it and handing it over. My freelance work for a garment exporter went exactly like that: from proposal and pricing to two shipped AI platforms and a handover.

I'm studying AI & ML at DJ Sanghvi alongside work, and I spend a fair few weekends at hackathons.

## Contact

- Email: himanshuk1205@gmail.com
- Location: Mumbai, India
- GitHub: https://github.com/Himanshu-Khairnar
- LinkedIn: https://www.linkedin.com/in/himanshu-khairnar-9908a6288
- Resume (PDF): https://himanshukhairnar.codes/Resume.pdf
- Website: https://himanshukhairnar.codes

## Experience

### Full-Stack & AI Engineer Intern — RankUp (Remote)
Jan 2026 – Present · current role

Building full-stack features around the UCP and ACP commerce protocols, and engineering the optimization and enrichment of Google Merchant Center (GMC) product fields to improve feed quality, coverage and listing performance.
- Built a UCP/ACP storefront on RankUp's commerce APIs — catalog, cart, passkey (WebAuthn) login and Razorpay checkout.
- Shipped ShopGPT as a Chrome side-panel extension so shoppers can chat, browse and check out without leaving the page.
- Engineered optimisation and enrichment of Google Merchant Center product fields to lift feed quality and coverage.
- Worked on an AI support agent with semantic product search, order tracking and returns for a Shopify brand.
- Built an MCP (Model Context Protocol) commerce agent that connects to Swiggy, Zepto and Zomato's MCP servers, letting shoppers search, cart and order in chat.
- Built a global payment wallet for AI shopping agents on Razorpay, with payment links and checkout sessions.
- Stack: Next.js, Python, UCP, ACP, Google Merchant Center

### Freelance Full-Stack & AI Developer — Tracon (Remote)
Apr 2026 – Aug 2026

Designed and shipped two AI platforms for a garment exporter: a Customer Intelligence dashboard that classifies the team's Gmail with Claude to surface buyer commitments, risks and SLA breaches, and PICC — a production command center with an AI agent over their SQL Server ERP. Owned scoping, proposals, delivery and handover.
- Stack: Next.js, Prisma, SQL Server, Supabase, Claude API, OpenAI, Gmail API

### Backend Developer & Database Manager — Softel Technologies Inc.
Jan 2025 – Aug 2025

Built internal CMS tools and database systems, wrote Python scripts to automate repetitive document operations, and contributed to backend APIs that sped up data processing for the Data Entry and Validation teams.
- Stack: Python, SQL, Node.js, Git

## Projects

- **RankUp — AI Search Optimization** (Work): Production SaaS that helps brands get found in AI answer engines: AEO query campaigns, AI-overview tracking, agent analytics, an AI-commerce product feed and the embeddable ShopGPT shopping assistant. Stack: Next.js, TypeScript, Supabase, Vercel AI SDK, LangChain, Firecrawl, MCP, Stripe. Links: [live](https://app.rankup.diy).
- **Tracon Customer Intelligence** (Client): Email-intelligence dashboard for a garment exporter — pulls the team's Gmail inboxes, classifies every message with Claude and surfaces buyer commitments, risks and SLA breaches, with daily digests and an AI chat agent. Stack: Next.js, Prisma, Supabase, Claude API, Gmail API, QStash, Recharts. Links: [live](https://tracon-dashboard.vercel.app).
- **Tracon PICC** (Client): Production Intelligence Command Center for a garment manufacturer — live views of factory output, PPC, QC, shipments and critical-path delays over their SQL Server ERP, with an AI agent that answers production questions in plain English. Stack: Next.js, Prisma, SQL Server, OpenAI, shadcn/ui, Tailwind. Links: [live](https://tracon-picc.vercel.app).
- **ShopGPT Storefront**: Conversational commerce on the UCP/ACP protocols — chat-driven product discovery, cart, passkey login and Razorpay checkout, shipped as a web storefront and a Chrome side-panel extension. Stack: Next.js, WebAuthn, Chrome MV3, Razorpay, Tailwind. Links: [live](https://acp-bice.vercel.app), [source](https://github.com/Himanshu-Khairnar/acp).
- **TypeArena**: Real-time multiplayer typing races with private rooms, friends, leaderboards, WPM analytics and per-key heatmaps. Stack: Next.js, Socket.io, Supabase, Zustand, Recharts. Links: [source](https://github.com/Himanshu-Khairnar/typingApp).
- **BillPartner**: Full-stack invoicing — create, email and track invoices with line items, tax slabs and discounts, PDF export and a revenue dashboard. Stack: Next.js, MongoDB, JWT, React-PDF, TanStack Table, Zod. Links: [source](https://github.com/Himanshu-Khairnar/Invoice_Generator).
- **PolarLink** (SIH 2026): Offline-first logistics for polar expeditions (MoES/NCPOR) — forecasts station survival time, models supply-ship delays and plans emergency evacuation routes. Stack: Next.js, FastAPI, PostgreSQL, NetworkX, D3-geo. Links: [live](https://polar-link-rho.vercel.app), [source](https://github.com/Himanshu-Khairnar/PolarLink).
- **Podcast Satya** (Team): Mobile app that extracts claims from health podcasts and fact-checks cancer-treatment misinformation with transformer models and an async processing pipeline. Stack: React Native, FastAPI, PostgreSQL, Redis, Transformers.
- **CodeClip**: Secure temporary clipboard — share text and files with a 4-digit code or QR, with expiry options and creator controls. Stack: Next.js, MongoDB, Cloudinary, QR. Links: [source](https://github.com/Himanshu-Khairnar/CodeClip).
- **Flowchart AI**: Turns plain-text descriptions into editable flowcharts with Gemini; syncs to a database and exports to PNG/PDF. Stack: Next.js, Gemini, React Flow, Supabase. Links: [source](https://github.com/Himanshu-Khairnar/flowchart-ai).

## Skills

TypeScript, JavaScript, Python, Java, C++, React, Next.js, Tailwind CSS, shadcn/ui, Node.js, Express, FastAPI, Flask, Django, Socket.io, PostgreSQL, MongoDB, MySQL, Redis, Supabase, Prisma, Firebase, Claude API, Gemini, LangChain, Docker, Git, GitHub, Linux, Vercel, Postman

## Education

- DJ Sanghvi College of Engineering — B.Tech, Artificial Intelligence & Machine Learning (2025 – 2028)
- Government Polytechnic, Mumbai — Diploma in Computer Engineering — 94.75% (2022 – 2025)

## Optional

- [GitHub profile](https://github.com/Himanshu-Khairnar): code and contribution history
- [Resume PDF](https://himanshukhairnar.codes/Resume.pdf): one-page resume