OnLearn

Your personal AI tutor landing page
Modules for DSA and Data Science(Will update further in future)
Your personalized learning plan. Updates dynamically as you progress.
This is your learning interface. The concept-tutor agent teaches pre-requisite concepts in a socratic approach.
Your integrated coding environment. The lab-mentor agent guides you throughout to write your code.
Your course structure to view the complete structure and navigate between the topics
Your progress tracker

Inspiration

While preparing for internships at IIT Roorkee, we found ourselves stuck in the same loop millions of students face—solving endless DSA problems, jumping between YouTube videos and one-size-fits-all articles, yet still feeling directionless.

According to our research:

4M+ engineering students in India struggle with technical placement preparation
Technical learners globally face the same “tutorial hell”

We realized the real issue isn’t the lack of content but the lack of personalized, real-time guidance.

That frustration, combined with our curiosity for agentic AI systems, inspired us to build the adaptive tutor we always wished we had.

What it does

OnLearn is an AI-powered adaptive tutor that identifies conceptual gaps, analyzes mistakes, and dynamically reroutes a learner’s path.

Key capabilities include:

Identifying conceptual gaps in a learner’s understanding
Analyzing mistakes and reasoning patterns
Dynamically rerouting learning paths based on mastery
Providing Socratic, conversational explanations
Adjusting difficulty based on pace and retention

Instead of acting like a static content platform, OnLearn behaves like a 1:1 intelligent mentor.

It bridges theory and coding practice through an integrated IDE, ensuring learners move from confusion to clarity through personalized learning pathways rather than one-size-fits-all content.

How we built it

OnLearn is built as a multi-agent AI learning system designed to orchestrate personalized learning journeys in real time.

Multi-Agent Learning Architecture

The system uses specialized agents that collaborate to guide a learner from concept understanding to coding practice.

Planner Agent
- Understands user goals during onboarding
- Generates personalized learning roadmaps for topics like DSA or Data Science
Master Agent (Retriever / Memory Manager)
- Maintains learning memory across sessions
- Retrieves past interactions and progress
- Orchestrates the overall learning flow
Concept Tutor Agent (Executor)
- Teaches concepts using Socratic explanations
- Guides learners step-by-step instead of giving direct answers
Lab Mentor Agent
- Evaluates coding solutions
- Provides debugging hints and feedback

This agent collaboration allows OnLearn to dynamically generate session tasks, guide reasoning, evaluate code, and update learning progress continuously. :contentReference[oaicite:0]{index=0}

System Architecture

The platform combines a modern web interface with a scalable AI orchestration backend.

Frontend

Next.js interface with an interactive dual-pane learning layout
Monaco Editor for integrated coding practice
Real-time chat interface for AI tutoring

Client-Side Execution

Pyodide (WebAssembly) runs Python directly in the browser
Enables zero-latency code execution without server round-trips

Backend Orchestration

FastAPI backend coordinates agent interactions and user sessions
Handles requests, context management, and task routing between agents

Agent Framework

Built using the DeepAgents framework
Used to manage:
- agent state
- tool access
- memory
- learning plan schemas

Infrastructure

MongoDB – stores user state, learning plans, and progress
Redis – caching and short-term session memory
Judge0 – secure sandboxed backend code execution

This architecture allows OnLearn to maintain persistent learning memory while delivering real-time tutoring and coding feedback. :contentReference[oaicite:1]{index=1}