AI Adaptive Learning Journey Generator

An AI-powered learning platform that assesses each student’s strengths and weaknesses, then creates a personalized learning journey with targeted lessons, practice, and continuous adaptation.

Comment

Inspiration

Students do not learn at the same pace or in the same way. However, traditional learning platforms often provide the same sequence of topics and resources to everyone. We wanted to create a system that understands what each student already knows, where they struggle, and what they should learn next. This inspired us to build AI Adaptive Learning Journey Generator.

What it does

The platform creates a personalized learning journey for each student.

  • Conducts a baseline assessment
  • Identifies topic-wise strengths and weaknesses
  • Calculates topic mastery
  • Generates a personalized learning path
  • Provides AI-generated notes and explanations
  • Recommends MCQs and coding practice
  • Provides English and Tamil learning resources
  • Reassesses students after learning
  • Dynamically adapts the next learning topics based on progress Core cycle: Assess → Analyze → Personalize → Learn → Practice → Reassess → Adapt ## How we built it We designed the system as an AI-powered web platform using:
  • Frontend: React, TypeScript, Tailwind CSS
  • Backend: Python, FastAPI
  • Database: PostgreSQL
  • Vector Search: pgvector
  • AI: Google Gemini
  • AI Architecture: AI Agents and RAG
  • Resources: YouTube Data API The system combines a predefined topic-prerequisite structure with AI-generated learning content. Topic mastery and prerequisites determine the student's learning progression, while AI generates personalized explanations, notes, quizzes, coding practice and resource recommendations. ## Challenges we ran into One of our biggest challenges was ensuring that AI-generated content was relevant to the student's actual learning needs rather than providing generic recommendations. We also had to design a reliable way to:
  • Measure topic-level mastery
  • Maintain prerequisite relationships between topics
  • Generate structured AI responses
  • Keep learning resources relevant
  • Dynamically update the learning journey after reassessment
  • Prevent the AI from overriding the defined learning logic ## Accomplishments that we're proud of We are proud of creating a learning approach that moves beyond a one-size-fits-all learning path. Our system can connect assessment performance with topic mastery and use that information to create a personalized journey. We also incorporated AI-generated learning materials, coding practice, quizzes and multilingual resources into the journey. Most importantly, the platform is designed as a continuous learning loop, where the student's progress changes what they learn next. ## What we learned We learned that building an AI application is not just about connecting an LLM to a website. We learned the importance of:
  • Structured data and topic relationships
  • Reliable AI outputs
  • Prompt engineering
  • Retrieval-Augmented Generation
  • Validation of AI-generated responses
  • Designing clear adaptive-learning logic
  • Combining deterministic rules with generative AI We also learned that AI should support decision-making, while important learning rules should remain controlled and predictable. ## What's next for AI Adaptive Learning Journey Generator Our next goal is to make the platform a more complete AI learning companion. We plan to add:
  • Multimodal learning from PDFs, slides, videos and other learning materials
  • A source-cited AI tutor
  • More advanced adaptive assessments
  • Personalized coding assistance
  • Deeper learning analytics
  • Better English and Tamil educational resources
  • Continuous adaptation based on long-term learning behaviour Ultimately, our vision is to move from one learning path for everyone to one intelligent learning journey for every student.

Built With

Share this project:

Updates