Inspiration

As students at the University of Washington, we realized how hard it can be to find the right peers to study or collaborate with — even when everyone around us shares similar goals.
We wanted to create something that bridges the gap between academics, interests, and community.
That’s how UW Connections was born — a platform that helps Huskies discover the best peers, courses, and study squads to grow together both academically and socially.

What it does

UW Connections is a personalized student networking and course recommendation platform.
It connects users based on shared classes, skills, and interests, helping them build study squads, track academic progress, and explore future career paths.
The platform offers:

  • Live course recommendations with progress and credit tracking
  • Study Squad feature to connect with peers
  • Skill progress visualization and gamified streaks
  • Career Path suggestions tailored to each student’s major and growth areas

How we built it

  • Frontend: React + Tailwind CSS for a responsive, minimalist interface
  • Backend: FastAPI server for course matching, recommendation, and user data APIs
  • Data: Real UW course data combined with simulated student interests in CSV format
  • Model: Hybrid recommendation system using cosine similarity

$$ \text{Score} = \alpha \cdot \text{CosineSim(Courses)} + (1 - \alpha) \cdot \text{CosineSim(Interests)} $$

where \$alpha$ adjusts the balance between academic and personal similarity.

  • Hosting: Uvicorn for local testing, with Docker setup planned for cloud deployment

Challenges we ran into

  • Integrating and cleaning large course datasets with inconsistent formatting
  • Ensuring recommendations were accurate yet diverse across different departments
  • Maintaining instant UI updates between My Courses, Study Squad, and Career Path tabs
  • Designing a personalized onboarding flow that stayed lightweight and intuitive

Accomplishments that we're proud of

  • Built a functional end-to-end matching system within a short time frame
  • Achieved an intuitive, gamified dashboard that keeps students motivated
  • Designed a recommendation algorithm that blends academic and social similarity
  • Created a prototype that could be extended to real UW student data in the future

What we learned

We learned how to:

  • Integrate machine learning into real-time web apps
  • Balance UI simplicity with backend complexity
  • Collaborate efficiently using Git, VS Code, and live API debugging
  • Turn student data into actionable, personalized insights that improve engagement

What's next for UW Connections Webpage

Looking forward, we plan to:

  • Add LinkedIn integration to import student skills
  • Expand career pathway recommendations based on academic performance
  • Develop a mobile-friendly Progressive Web App (PWA)
  • Introduce an AI academic assistant chatbot to help new students navigate courses and communities
  • Deploy the platform publicly to help more UW students connect, learn, and grow.4444fgvvc fgf
Share this project:

Updates