Ramify

Inspiration

Finding the right people to collaborate with on campus can be difficult. Many students want guidance, research opportunities, or mentorship, but information about professors, alumni, and clubs is often scattered across so many different websites. As a result, students, especially first-years, are overwhelmed with the amount of information.

What it does

Our platform lets students create a simple profile with their year, department, interests, and goals. It then recommends professors, alumni, or campus groups that are a strong match. Each connection is ranked by a fit score so students can quickly find the people most relevant to their academic or career interests.

How we built it

We built a lightweight web app where students enter their academic information, interests, and goals. The backend processes this information and uses Amazon Bedrock with the Claude 3 Haiku model to generate personalized match explanations. The system ranks professors, alumni, and campus groups based on how well they align with the student’s profile. The interface then displays the top matches along with reasons why they are a good fit, making it easier for students to discover relevant connections.

Challenges we ran into

One challenge was designing a matching system that actually produced meaningful recommendations instead of generic results. We had to experiment with how we structured user inputs and prompts so the AI could generate useful explanations for each match. Another challenge was integrating the Bedrock API with our application and ensuring the responses were fast enough to keep the user experience smooth.

Accomplishments that we're proud of

We’re proud that we were able to turn an idea into a working platform within a short hackathon timeframe, and especially as first years. The system successfully connects student profiles to relevant professors, alumni, and campus groups, and generates clear explanations for why each match makes sense. We’re also proud of integrating Amazon Bedrock to power the recommendation logic, which allowed us to build a smarter matching system rather than relying on simple keyword searches.

What we learned

Through this project, we learned how to integrate AI models into a real application workflow. We gained experience working with the Amazon Bedrock API and Amazon Q, designing prompts for meaningful responses, and connecting AI outputs to a user-facing interface. We also learned how important it is to structure user data clearly so the system can generate useful and relevant recommendations.

What's next for Ramify

Next, we want to expand the platform by adding more professors, alumni, and student organizations using the school's public data, so the recommendations become even more useful. We would also like to integrate official university directories and research listings to keep information up to date automatically. In the future, Ramify could include messaging features, event recommendations, and mentorship matching so students can not only discover connections but also start meaningful conversations and collaborations directly through the platform.