LearnFlow-AI

• 

  landing page

• 

  academics dashboard

• 

  side hustle dashboard

• 

  roadmaps

• 

  agent timeline

• 

  insights + skill gap analyzer

• 

  insights page

Inspiration Students today want more than just good grades — they want career-ready skills, strong portfolios, side hustles, and real-world impact. However, most learning systems are fragmented and reactive, forcing students to constantly guess what to learn next and how to stay consistent. Many start exam preparation late, drop off while learning skills, and struggle to convert effort into meaningful outcomes. We built LearnFlow AI to address this gap by creating an intelligent system that doesn’t just provide content, but actively guides direction, prioritization, and long-term growth, helping students turn scattered effort into structured academic and career success.

What it does LearnFlow AI is a multi-agent AI platform that functions as a long-term mentor for students, helping them build career skills while managing academics intelligently. The system continuously analyzes goals, deadlines, behavior, and progress to guide users on what skills to learn next, how to close skill gaps, and how to convert learning into real projects and portfolios. It integrates resume building, project evaluation, and adaptive study planning into a single ecosystem, ensuring that students stay consistent with side hustles while remaining academically prepared. Instead of static planning tools, LearnFlow AI adapts in real time, evolving with the student’s journey.

How we built it We built the platform using a Next.js frontend for dashboards, onboarding, insights, and timelines, paired with a FastAPI backend that handles authentication, exams, assignments, profiling, and AI-driven routes. At the core is a multi-agent architecture composed of planning, urgency, progress, context, exam parsing, and career intelligence agents working together. Firebase/Firestore stores user data, while RAG-based AI pipelines power tutoring, skill analysis, and adaptive recommendations. This architecture allows the system to maintain long-term context and deliver personalized decisions rather than isolated responses.

One of the biggest challenges was orchestrating multiple AI agents while maintaining a unified context of the student’s behavior. Designing adaptive logic that produces meaningful career guidance — rather than generic AI output — required careful experimentation. We also worked extensively on optimizing heavy AI routes, handling long-term memory, and structuring a backend architecture that could scale with growing complexity.

Accomplishments We are proud to have built a career-first AI learning system, not just another planner. The integration of skill gap analysis, resume generation, evaluated projects, RAG tutoring, and adaptive academic planning demonstrates a real multi-agent architecture in action. Most importantly, the product shifts focus from task completion to measurable career outcomes, helping students build proof of skills rather than just checklists.

What we learned Through this project, we learned how to design scalable AI-driven systems, orchestrate specialized agents, and build adaptive recommendation pipelines. We also gained a deeper understanding of student consistency challenges and how AI can act as a mentor that evolves with the user. Balancing academic structure with career growth in a single platform taught us how intelligent systems can support real human behavior.

What’s next Our next steps include adding AI-powered portfolio builders, real-time project collaboration, internship matching, mobile apps, and voice-based mentoring. The long-term vision is to evolve LearnFlow AI into a full career co-pilot, guiding students from college learning to industry readiness with continuous adaptive intelligence.

Built With

• fastapi
• firebase
• langgraph
• nextjs
• python
• typescript