Hackathon Summary
Inspiration
Driven by the belief that quality education should be universally accessible, Learnbridge was inspired by the challenges faced by learners in under-resourced communities. We set out to create a social-good platform that automates course creation and support, lowering barriers to knowledge.
What it does
Learnbridge transforms a simple syllabus or outline into a full-featured online course. Users can generate modules and lessons automatically, enriched with YouTube videos and web-sourced citations. Additionally, a personal chatbot assistant provides contextual help based on the current lesson, enabling interactive, self-paced learning.
How we built it
- Backend: Flask API with Celery for asynchronous lesson generation tasks
- Orchestration: Sequential multi-step agent pipeline (LangGraph) to decompose and coordinate content creation
- Frontend: Next.js with Material UI for a responsive, user-friendly interface
- Auth & Storage: Firebase Authentication for secure access
Challenges we ran into
- Timeouts & Memory Limits: Initial monolithic generation caused failures; solved by splitting tasks with Celery
- Agent Coordination: Designing a reliable orchestration pipeline required extensive prompt engineering and debugging
- Content Quality: Ensuring relevance and citation accuracy in retrieved web and video resources
Accomplishments that we're proud of
- Built a fully autonomous course-generation pipeline with reliable background processing
- Integrated contextual chatbot that retains lesson context and memory for meaningful assistance.
- Delivered a working prototype within hackathon timeframe
What we learned
- The importance of task isolation and asynchronous processing for scalability
- Effective prompt design and chaining techniques for multi-step agent workflows
- Strategies for integrating external retrieval tools (YouTube, web search) into AI-driven apps
What’s next for Learnbridge
- UX Enhancements: Add progress indicators, loading states, and improved error handling
- New Features: Audio/video note creation, “learn mode” with spaced repetition, course analytics
- Expanded Tooling: Incorporate additional knowledge sources and multi-language support
- Community Building: Open-source roadmap and invite educators to contribute syllabi templates
Log in or sign up for Devpost to join the conversation.