AI Adaptive Learning Journey Generator

• 

  file

• 

  resourse

• 

  course avaliable

• 

  learning page

• 

  home

Inspiration

Traditional learning platforms often provide the same learning path to every student, even though students have different strengths, weaknesses, and learning speeds. We wanted to build an AI-powered learning companion that understands a student's learning materials and creates a personalized journey based on their actual performance.

What it does

Our Adaptive Learner accepts multimodal learning resources such as PDFs, slides, notes, and lecture content. The system analyzes the material, builds a source-cited knowledge base, and assesses the student's topic-wise understanding. Based on the results, it identifies weak and mastered topics and generates a personalized learning path with explanations, notes, quizzes, coding practice, and AI tutoring. After reassessment, the learning journey dynamically adapts to the student's progress.

How we built it

We built the platform using React and Tailwind CSS for the frontend and Python/FastAPI for the backend. PostgreSQL with pgvector is used for storing learner data and enabling semantic retrieval. Gemini processes learning content, generates assessments and explanations, and supports personalized recommendations. The system follows an adaptive cycle: Assess → Analyze → Plan → Learn → Practice → Reassess → Adapt.

Challenges we ran into

One of our main challenges was ensuring that AI-generated content remained relevant to the student's actual learning material instead of producing unrelated answers. We addressed this using retrieval-based generation and source citations. Another challenge was designing a reliable topic-wise mastery system so that the AI recommendations were based on measurable student performance.

Accomplishments that we're proud of

We are proud of building an adaptive learning workflow that goes beyond a simple AI chatbot. The system can understand learning resources, identify individual knowledge gaps, generate personalized learning journeys, and continuously update recommendations based on assessment results. We are especially proud of combining multimodal AI, adaptive assessment, RAG, and personalized tutoring into one learning platform

What we learned

We learned that building an effective AI education system requires more than connecting an LLM to a chatbot. Reliable retrieval, structured AI outputs, topic-wise assessment, source grounding, and deterministic learning rules are essential for producing useful and trustworthy personalized learning experiences.

What's next for AI Adaptive Learning Journey Generator

Our next goal is to make AI Adaptive Learning Journey Generator a complete multimodal AI learning companion. We plan to expand support for videos, PDFs, presentations, and images, improve source-cited RAG, and introduce a more advanced AI tutor that can explain concepts interactively. We also plan to improve personalized assessments, coding practice, progress analytics, and continuous learning-path adaptation based on each student's performance.

Built With

• agents
• ai
• api
• css
• data
• database
• fastapi
• gemini
• generative
• multimodal
• pgvector
• postgresql
• python
• rag
• react
• tailwind
• typescript
• vector
• youtube