Inspiration
Modern education is broken in a subtle but critical way โ students consume content, but they rarely internalize it. Notes are static. Classes are one-directional. Practice is disconnected.
We wanted to build something fundamentally different: ๐ a system that understands how you learn, adapts to your mistakes, and guides you like a real mentor.
Exceed was inspired by a simple question:
What if every student had a personalized AI system that tracks their learning journey, understands their weaknesses, and actively helps them improve?
What it does
Exceed is an AI-powered adaptive learning ecosystem, not just another edtech tool. It connects everything:
๐ Notes โ converted into structured learning journeys ๐ฎ Practice โ gamified, adaptive, and engaging ๐งโ๐ซ Tutor โ AI that teaches, tests, and adapts ๐ Planner โ dynamically adjusts based on your mistakes ๐ซ Classes โ real teacher-student collaboration with analytics ๐ฅ Key Features ๐งฉ Concept Journey (Story Mode)
Any uploaded note transforms into an interactive learning path:
Concepts broken into step-by-step nodes Locked progression (no skipping โ mastery required) Quiz checkpoints at every stage Completion unlocks achievement badges
๐ Learning feels like progression, not passive reading.
๐ฏ Smart Practice Engine Generates questions from your own notes Adapts difficulty in real-time Includes: Quiz mode Flashcards Story-based application Runner-style game ๐ฎ
๐ Every mistake is tracked and used to improve you.
๐ง AI Tutor (Context-Aware) Teaches from your own content Explains, asks, evaluates, and adapts Voice-enabled + interactive learning flow
๐ Not a chatbot โ a true learning guide.
๐ Intelligent Study Planner Builds a study plan based on: weak concepts mistakes deadlines Uses spaced repetition + workload balancing
๐ You donโt plan your study โ the system plans it for you.
๐ซ Class System + Teacher Analytics Students join via class code Access sessions, notes, and materials Teachers get: mistake analytics engagement tracking concept-level performance
๐ Teaching becomes data-driven and precise.
๐ Meaningful Leaderboards Not based on activity โ based on learning quality Encourages growth, not just competition
How we built it
We designed Exceed as a modular AI-driven system:
๐ง Learning Intelligence Engine Tracks every attempt, mistake, and concept mastery ๐ RAG Pipeline Converts notes โ structured knowledge โ questions โ journeys ๐ฎ Gamified Frontend Built with modular UI and smooth interactions ๐๏ธ Unified Data Layer Connects Practice, Tutor, Planner, and Classes
We also ensured:
โก offline support (via local models like Ollama) ๐ scalable backend APIs ๐จ clean, production-grade UI
Challenges we ran into
Designing a system where everything connects meaningfully Avoiding fake gamification and building real learning impact Creating smooth, performant interactions (especially in game modes)
Accomplishments that we're proud of
Built a fully connected learning ecosystem (not isolated features) Replaced mock metrics with real learning intelligence Created a Concept Journey system that feels like progression Designed a planner that actually adapts to student performance Combined AI, UX, and education into a cohesive product
What we learned
Learning is not about content โ itโs about feedback loops Gamification works only when tied to real progress AI is powerful, but only when structured into clear systems
What's next for Exceed
๐ Deeper personalization using long-term learning patterns ๐งโ๐ซ Teacher AI assistants for content generation & insights ๐ Advanced predictive analytics (exam readiness, risk detection)
๐ฏ Final Vision
Exceed is not just a platform.
Itโs a system where:
every mistake matters every concept is tracked every student gets a personalized path
๐ Weโre building the future of adaptive, intelligent learning.
Built With
- node.js-(express)
- ollama-(local-llm)
- openai-api
- postgresql
- react.js
- tailwind-css
- typescript
- websockets
Log in or sign up for Devpost to join the conversation.