AI Engineer · React & Next.js Architect · SaaS Builder · Open Source Developer
Core developer at @ever-co · Ahmedabad, India (IST, UTC+5:30)
Website · LinkedIn · Email · X · Upwork
I build products end to end — the AI layer, the front end, the API, and the platform underneath — and ship them as inspectable work: a repository, a demo, or a stated status.
Most of my work sits in the TypeScript ecosystem. React and Next.js on the product side, Angular on the platforms I maintain, NestJS and Node.js behind them. I spend my time on systems that have to hold up under real use — ERP, CRM and HRM — and on the AI tooling that makes building them faster.
| 🤖 AI applications | LLM-backed product work — the direction the rest of this list serves |
| 🔌 MCP servers | Model Context Protocol tooling for AI development workflows |
| ⚛️ React 19 & Next.js | Product front ends and statically exported sites, on a real performance budget |
| 🛠️ AI developer tools | The tooling layer around building with models, not just calling them |
| 🚀 Full stack SaaS | Multi-tenant products, from schema to shipped UI |
| 🏢 Ever Gauzy | Core development on an open-source ERP/CRM/HRM platform |
| Project | Role |
|---|---|
| Ever Gauzy — open-source ERP/CRM/HRM platform | Core developer |
| Ever Gauzy Teams — work & project management platform | Backend API developer |
| ngx-daterangepicker-material | UI fix (empty ranges) |
| nestjs-multi-orm — dual-ORM NestJS module | Author |
Organized by domain to highlight the technologies I use most in modern AI-powered product development.
| Category | Technologies |
|---|---|
| Assistants | Claude Code, ChatGPT, GitHub Copilot, Google Gemini |
| SDKs | Anthropic API, OpenAI API, Vercel AI SDK |
| Frameworks | Model Context Protocol (MCP), LangChain |
| Category | Technologies |
|---|---|
| Frameworks | React, Next.js, Angular |
| Design systems | Tailwind CSS, shadcn/ui, Radix UI, Material UI, Angular Material, Framer Motion |
| State & data | React Query, TanStack Table |
| Languages | TypeScript, JavaScript, HTML5, CSS3 |
| Category | Technologies |
|---|---|
| Node.js | Node.js, NestJS, Express.js |
| PHP | Laravel, Lumen, CodeIgniter |
| APIs | REST, GraphQL, WebSockets |
| Integrations | Stripe, Razorpay, OAuth, Firebase Services, third-party APIs |
| Category | Technologies |
|---|---|
| Databases | PostgreSQL, MySQL, MongoDB, SQLite |
| Data access | Prisma ORM, Drizzle ORM, TypeORM, Supabase Client |
| Category | Technologies |
|---|---|
| Cloud | AWS, Cloudflare, Vercel, DigitalOcean, Linux |
| BaaS | Supabase, Firebase |
| Containers | Docker |
| CI/CD | GitHub Actions, GitLab CI/CD |
| Deployment | Vercel Deployments, Cloudflare Pages |
| Category | Technologies |
|---|---|
| Build | pnpm, Turborepo, Nx, ESLint, Prettier |
| Quality | Vitest, Jest, Playwright, Cypress, Testing Library |
| Observability | PostHog, Sentry, Google Analytics, Google Tag Manager |
| Tools | Git, GitHub, GitLab, VS Code, Postman |
Modern software engineering is no longer about writing every line of code by hand. I use AI across the whole development lifecycle to ship higher-quality products faster — as a collaborative engineering partner, not as a replacement for engineering judgement.
| Stage | What AI does for me |
|---|---|
| 💡 Product discovery | Turn ideas into technical specifications and PRDs, research competitors and market trends, validate architecture decisions, break large initiatives into milestones |
| 🏗️ System design | Design scalable architectures, define API contracts, model databases, write ADRs, generate sequence and architecture diagrams |
| 💻 Development | React & Next.js applications, TypeScript-first code, Node.js & NestJS APIs, AI-powered features, refactoring and modernisation, test generation, performance work |
I pick the right tool per task rather than relying on a single assistant: Claude Code, ChatGPT, GitHub Copilot, Gemini, Cursor and Windsurf for the assistant layer; OpenAI and Anthropic APIs, the Vercel AI SDK and Model Context Protocol (MCP) for what I build with; local AI tooling where it fits.
Idea → Research → Product requirements → Architecture → Implementation → Code review → Testing → Optimization → Deployment → Monitoring
AI accelerates every stage. Validation, engineering judgement and the final implementation stay mine — AI accelerates development, not decision making.
I build AI-powered products, developer tools, and SaaS applications with React, Next.js, Node.js and TypeScript. Clean architecture and developer experience matter as much as shipping features — a codebase that is pleasant to work in is the reason the second year of a product goes faster than the first, not slower.
Three things I hold to:
- Ship inspectable work. A repository, a demo, or an honest status. Claims without an artifact behind them are just claims.
- Type the boundaries. Most production bugs are contract bugs. TypeScript at every seam is the cheapest defence there is.
- Budget for the reader. Performance budgets, tests and accessibility checks are not polish added at the end; they are what stops a product degrading with every release.
The header carries the same links as a compact row; this table says what each one is actually for, so the two are not simply a repeat of one another.
| Channel | Best for |
|---|---|
| 🌐 Website | Longer-form background and the professional record |
| Roles, introductions, and anything hiring-related | |
| Direct enquiries and anything that needs a written reply | |
| 🐦 X | Short-form updates on what I am building |
| 💬 GitHub | Issues, pull requests, and questions about the code |
| 🤝 Upwork | Contract and freelance engagements |
| 📍 Ahmedabad, India | Timezone IST (UTC+5:30) |
I take on product work where the architecture matters as much as the feature list. My repositories are open — read the code first, then get in touch if it raises a question worth answering.