I build and ship AI products that real people use — voice platforms, campus assistants, personal AI tools. If you're working on something with AI, I'd love to hear about it.
I'm an AI Developer & Product Builder at Jaipuria Institute of Management, where I design and ship LLM-powered products — Rehearsal AI, JChat, and life.ai among them. Most of my days go into turning ideas into tools that students and faculty can genuinely use.
I try to stay involved end to end. On one side, the AI itself — Claude and GPT, RAG, and voice. On the other, the product, the analytics, and the DevOps I manage to keep each release tested, secure, and running. I'm still learning plenty as I go, and I'm always happy to talk through anything AI-related.
A voice-based mock interview platform built on Next.js, Claude, and ElevenLabs, with real-time AI feedback on how you actually performed. 3,000+ students use it, and PostHog analytics across the product guide what gets built next.
A personal AI assistant that helps you manage tasks, workflows, and the small frictions of a working day through plain conversation. Building it as a standalone web product, full experience designed end to end.
A reading tool that keeps the intent behind every highlight — colour-coded meaning and your own connections between ideas, instead of flattening everything to plain text. It's built around the moment you decide something matters. Live at highlyt.app.
A white-labelled AI assistant for Jaipuria Institute that draws on multiple LLMs to answer students and faculty in context. Built, launched, and live for the whole campus.
A GPT model built layer by layer — ~12 million parameters trained on a custom Bhagavad Gita dataset, ~97% accuracy. The project that taught me what's actually under the hood of every LLM I now build with.
Built and shipped Rehearsal AI and JChat — voice and chat products now used by 3,000+ students — owning them from first idea through launch. Designed the analytics that guide product decisions, and managed the DevOps that keeps each release tested, secure, and steady in production.
Sessions for MBA students, faculty, and HR teams on using AI at work — Claude, ChatGPT, RAG, and where to be careful. Practical and hands-on, built around what people can actually use the next day.
Built and shipped REST APIs for referral and brand verification into the live system — production-stable and backward-compatible.