This is a submission for the Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine.
What I Built
I built Cognivanta, an AI-powered learning platform that turns study notes into interactive learning experiences.
The idea behind the project is simple: every set of notes contains ideas waiting to be connected. Cognivanta helps bring those ideas to the surface by transforming uploaded study material into structured concepts, explanations, concept relationships, and quiz questions. Instead of only rereading a page, learners can explore what it means, see how its ideas relate, and test what they remember.
I interpreted “Ghost in the Machine” as the hidden knowledge inside our everyday learning materials: the concepts and connections that are present in a page of notes but not immediately visible. Cognivanta uses AI to help reveal that “ghost” and turn it into something learners can explore.
The project also includes interactive subject experiences for Biology, Physics, History, and Geography, alongside profile and learning-progress features.
Demo
Video demo: https://youtu.be/22bTe2ij43c?si=3JTUehrf7DtWUg-h
The video will walk through the core experience: uploading study material, generating learning content, exploring connected concepts, and trying an interactive subject world.
Code
GitHub repository: Rohi56u/COGNI-VANTA https://github.com/Rohi56u/COGNI-VANTA
Cognivanta is built as a multi-service application, with a React and Vite frontend, an Express backend, and a Python FastAPI service for AI-assisted content generation.
How I Built It
The frontend is built with React and Vite. It provides the learning workspace, upload flow, concept visualizations, quizzes, subject experiences, and profile views.
The backend uses Node.js, Express, and SQLite for authentication, document handling, learning data, and progress-related features. A separate Python FastAPI service processes uploaded material and connects it to Gemini for AI-assisted transcription and concept generation.
The AI workflow is designed to turn study material into structured learning content, including concepts, relationships, and recall questions. The repository also includes a sample-data path for development and demonstration.
Why Does Open Innovation Matter?
Cognivanta is open source because learning tools should be open to inspection, adaptation, and improvement. Developers and educators can explore how the application is put together, adapt the learning experience for different subjects, and contribute ideas or features of their own.
The project currently uses Gemini’s API for AI generation, so I’m not presenting the model itself as open source. Open innovation still matters here: the surrounding application, learning flows, and integration are available for others to study and build on. Keeping those parts open also makes it easier to experiment with alternative models and providers over time.
My Agent Session
[Optional: Add your DevRelay session link or embed here.]
The Spirit of It
A page of notes can look like a collection of disconnected facts. But behind it are ideas, patterns, and questions a learner may not have noticed yet.
Cognivanta is my attempt to make that hidden layer visible: to turn notes into connections, connections into questions, and questions into a more active way of learning.
Top comments (0)