I'm Saurabh Jaykar: Full Stack Engineer Building AI-Powered Products

I'm Saurabh Jaykar, a full stack engineer based in Mumbai, India. I build web and mobile products with React, Next.js, Node.js, and Python, with a focus on AI workflows, backend systems, payments, and cloud deployments.

I have more than two and a half years of experience shipping software. I enjoy working across a product: understanding what someone needs, building the interface, designing the APIs behind it, and getting the complete experience into production. This is an introduction to the work I do, the products I've contributed to, and what interests me as an engineer.

What I work on at Vedic Rishi Astro

Since August 2025, I've worked as a Full Stack Engineer at Vedic Rishi Astro in Mumbai. My work brings together astrology products, AI-assisted website creation, backend modules, payments, and campaign automation. These projects give me opportunities to work on both the parts users interact with and the systems supporting them.

One of those products is AstroPages, an AI-powered no-code website builder for astrologers. It enables people to create, customize, preview, and publish production-ready websites for their practice. I contributed to the product by delivering and maintaining 24 reusable Astro templates and Cloudflare deployment components.

Reusable templates make consistency important: a change needs to work across different content and page layouts. I like that combination of frontend detail and platform thinking. It involves helping individual websites feel complete while keeping their shared building blocks maintainable.

Backend ownership and AI Builder workflows

At Vedic Rishi, I owned backend development for the Project Leads and Payment Currency Preference modules. That included role-based REST APIs, INR/USD currency logic, and concurrency-safe updates. These features need clear rules around who can change data and how simultaneous updates are handled.

I also worked on Python/FastAPI AI Builder workflows using reusable skills, MCP integrations, and prompt engineering. My interest in AI engineering is practical: connecting model capabilities with the tools and product workflows that make them useful. The surrounding APIs, permissions, and deployment process are part of that work.

My toolkit includes Node.js and Express for backend development, Python and FastAPI for AI-related workflows, and databases such as PostgreSQL, MongoDB, and DynamoDB. I choose tools based on what the feature needs and how it will be maintained.

UpAstro and campaign automation

I built UpAstro, a western astrology platform using Next.js and Cloudflare, with Stripe and PayPal subscriptions and astrology reports. I also developed an AI-powered couple-compatibility platform using astrology APIs. Together, these astrology experiences served 2,000+ international users.

Building these products involved connecting the user experience with payments, external APIs, and generated reports. It reinforced how much product quality depends on the connections between features: a polished screen is only one part of a complete workflow.

Another part of my role was building campaign automation for WhatsApp, email, and push notifications. User segmentation helped increase campaign open rates by 28%. I integrated Amazon SES tracking for delivery, bounce, and spam metrics so campaign performance could be understood beyond whether a message had been sent.

My earlier work on Propfusion CRM

Before Vedic Rishi, I worked at AffWorld Technologies from January 2024 to July 2025. I developed Propfusion CRM, an AI-powered real estate CRM adopted by 20 Dubai firms and 1,000+ agents. The product brought together property listings, document management, and in-app chat.

I integrated the Gemini API for automated lead follow-ups and generated property descriptions. I also implemented JWT-based authentication with role-based access control and optimized database queries to reduce API response times.

Working on a CRM taught me to think carefully about shared data, different user roles, and the everyday workflows a team depends on. Those concerns continue to shape how I approach backend development and product features.

The projects I build outside work

I publish and maintain personal projects alongside my professional work. toast-ninja is a lightweight, zero-dependency React notification library published on npm. It supports TypeScript, a hooks-based API, and smooth animations, with documentation deployed on Vercel. Building a library makes the developer experience part of the product.

Task Buddy is a MERN task-management application with JWT authentication, task operations, real-time updates, and filters for priority and progress. Its interface uses React 19, Tailwind CSS, and Vite.

QuickLinks is a URL shortener with a React client and a TypeScript, Express, and MongoDB backend. It includes authenticated link management and click analytics. I also work on GuideDocs, a documentation project for people learning development. You can explore more of these projects on my portfolio and GitHub.

How I approach engineering

I care about clean architecture, useful interfaces, and understanding how a feature behaves after it is deployed. I prefer reusable components where they reduce repeated work, clear API boundaries, and straightforward solutions that another developer can understand.

I enjoy owning features from idea to production and learning from the constraints along the way. My Bachelor of Commerce at Savitribai Phule Pune University and my full-stack development training are part of my background; building and maintaining real products continues to be an important part of how I learn.

Let's connect

I'm open to full-stack and AI product engineering opportunities. If you're hiring someone who can work across React and Next.js frontends, Node.js and Python backends, and the integrations connecting them, I'd be happy to talk.

You can email me, connect on LinkedIn, or follow my work on GitHub and X. My work experience has more details about the products and systems I've helped build.