Hello everyone,
I’m excited to share a project I’m currently in the early stages of building: Kalpavriksh, an AI-powered learning platform that “grows a knowledge tree” to explain any concept.
The vision is to go beyond simple text answers. It takes a user’s input (like “Newton’s First Law”) and will eventually generate a complete “branch” with three key parts:
-
Explanation: A simple, concise breakdown.
-
Real-life Examples: Practical, everyday demonstrations.
-
Simulation: A structured JSON output that powers an interactive animation in the UI.
I’m currently building the backend using LangChain with OpenAI’s models to orchestrate these different outputs. The core challenge I’m focused on right now is designing the right prompts and chaining logic to get reliable, structured output, especially for the simulation part.
I’m still very much in the prototyping phase and don’t have a live demo or a public repo yet, but I wanted to share the idea and get some early insights from this community.
I’d love to hear your thoughts on this approach. Do you have any experience generating structured, multi-step output for visualizations? Any tips on designing prompts for this kind of specific, chained output would be incredibly helpful as I continue to build.
Thanks! I’ll be sure to post an update once I have a working prototype to share.