π Mumbai, India | π§± Senior Frontend Engineer | π€ AI-First Builder | π Indie SaaS Maker
8+ years building scalable frontend systems β promoted through four roles at AccionLabs. Shipped features to 14M+ users, contributed to a $9M revenue increase, and won the Innovation & Engineering Excellence award at the 2026 AccionLabs Summit.
I architect component systems, establish engineering standards, and lead AI-first workflows. Outside of work, I build and ship indie SaaS products end-to-end.
| Project | Description | Stack | |
|---|---|---|---|
| weavr | Rust CLI that converts Claude Code transcripts into beautiful self-contained HTML β 18β46Γ faster than the Python alternative. Published on crates.io and Homebrew | Rust SQLite Material 3 |
|
| LexLight | AI-powered legal document analysis β plain-English contract review with risk scoring, clause-by-clause breakdown, and PDF/Markdown export. Lighthouse score of 100 | Next.js 16 React 19 TypeScript AI SDK |
|
| Claude Desk | Browser UI for managing Claude Code config β edit skills, plans, agents, and settings without hand-editing JSON. One command: npx claude-desk |
React Hono TypeScript Vite+ |
|
| Requestr | Browser-native API testing tool β Postman alternative with zero backend, IndexedDB persistence, and full collection support | React IndexedDB PWA |
|
| Visual AI | AI SaaS for image generation, upscaling, colorization & restoration β built solo, end-to-end | Next.js AI SDK Cloudinary |
|
| Elementum | Interactive periodic table with 3D atom visualisations, heatmaps, and Lighthouse score of 100 | Three.js React TypeScript |
|
| CreateFolio | Portfolio generator with one-click Netlify deployment | React Netlify Serverless |
|
| RocketLander | Physics-based browser game with Firebase leaderboard at a steady 60 FPS | Phaser Firebase Canvas |
Frontend
React Next.js Vue 3 Angular TypeScript Tailwind ShadCN MUI Ant Design
State & Data
Redux Pinia Jotai TanStack Query GraphQL REST
Testing
Jest React Testing Library Vitest Cypress Playwright
Backend & Systems
Node.js Express Hono Rust MongoDB Redis Docker GitHub Actions CI/CD
AI & Tooling
Claude Code GitHub Copilot Cursor Prompt Engineering AI SDK
Platforms
Vercel Firebase Netlify DigitalOcean Cloudinary
I introduced and scaled AI-assisted workflows across my team β cutting sprint delivery time by ~30%. Day-to-day this means:
- Reusable prompt templates for tests, code reviews, and commit messages
- Using Claude Code and Copilot for architectural decisions, not just autocomplete
- Building internal tooling that makes the whole team faster
"I care about the last 10% as much as the first 90% β that's where good engineering becomes great product."

