I Built Kikar — An AI Messaging Platform That Creates Digital Versions of People

Developer Bhavya4907 has built Kikar, an AI-powered social messaging platform that creates digital versions of users, learning from conversations to chat in their style. The platform uses Next.js, React, TypeScript, Tailwind CSS, Supabase, Capacitor, and the Groq API, and is available on GitHub. The developer plans to improve AI memory, add voice features, and enhance UI.

For the last few months, I've been working on Kikar, an AI-powered social messaging platform with a simple goal: What if every user could have an AI version of themselves that learns from conversations and can chat in their own style? Instead of building another generic chatbot, I wanted to explore the idea of AI personalities that evolve over time. Tech Stack Next.js React TypeScript Tailwind CSS Supabase Capacitor Groq API The biggest thing this project taught me wasn't AI—it was how all the pieces of a real application fit together. From authentication and databases to APIs and frontend architecture, I learned far more than I expected when I first created the repository. There's still a long way to go. I want to improve AI memory, add voice features, polish the UI, and make the conversations feel even more natural. If you have a few minutes, I'd genuinely appreciate any feedback—whether it's about the code, the design, or the idea itself. GitHub: https://github.com/Bhavya4907/Kikar https://github.com/Bhavya4907/Kikar %} Thanks for reading. I hope this project keeps getting better with every update.