Inspiration Students rely on separate tools for parking, dining, events, projects, and careers. We wanted to bring everything into one intelligent platform. What it does HokieFlow helps Virginia Tech students make smarter decisions across campus life and career growth. It recommends parking, predicts dining rushes, suggests study spaces and events, manages projects, matches students with jobs, tailors resumes, and prepares applications. How we built it We designed HokieFlow around a multi-agent AI system, with specialized agents for parking, dining, careers, projects, resumes, and applications. Databricks serves as the planned data and intelligence layer, while Demo Mode keeps the prototype reliable when live integrations are unavailable. Challenges The biggest challenge was combining many student needs into one cohesive experience. We solved this by focusing every feature on one question: “What is the smartest next move for this student?” What we learned AI is most useful when it combines data, context, and specialized agents to make actionable recommendations—not just answer questions. What’s next Next, we want to connect HokieFlow to live Virginia Tech systems for parking, dining, transit, schedules, events, and career opportunities and also expand it to different universities in the future .
Built With
- agents
- ai
- apis
- catalog
- css
- databricks
- fastapi
- leaflet.js
- llms
- mlflow
- multi-agent
- next.js
- openstreetmap
- python
- react
- rest
- search
- sql
- systems
- tailwind
- typescript
- unity
- vector
Log in or sign up for Devpost to join the conversation.