GuidEd

GuidED: The AI-powered platform helping students navigate college smarter—connecting them to the right opportunities, resources, and peers.

Comment

Inspiration

GuidED was inspired by the challenge students like us face in finding mentorship, funding, and opportunities in large, decentralized institutions. By leveraging AI and community-driven connections, it bridges the gap between ambition and access, ensuring every student can unlock their full potential.

What it does

GuidED, is an AI-powered, resource-matching platform designed to help students find exactly what they need—when they need it. Through partnerships with universities, GuidED systematically onboards students, learns about their goals, and matches them with the right resources at the right time—whether it’s funding, mentorship, events, or networking opportunities.

How we built it

We used mainly react and javascript to build our app. We created a firebase database to store profile information on a student and information about resources available in the entrepreneurship center at brown. Then, we used ChatGPT’s API to create a chatbot interface that can provide targetted recommendations for the student.

Challenges we ran into

We were hoping to fully integrate our app with AWS services such as DynamoDB, BedRock, Lex and Knowledge Bases and test our AI with different models such as Llama and Claude. However, our biggest challenging was setting up the permissions and connections correctly. However, we learned so much about AWS along the way even though we were not able to fully integrate it. ** Despite the challenges, we do use AWS Amplify to host our entire project!** Moreover, we spent quite a bit of time figuring out how to integrate the frontend and backend code. Especially, on how to connect to a database API at the same time as an OpenAI API and ensure information is received correctly (ie. we ran into issues where we had no data to use in our prompt because Firebase hadn't finished querying yet)

Fine tuning our parameters, prompt, and testing the AI responses was also quite difficult, as it was difficult to think of some edge cases to make sure the AI returned correct information with the correct tone.

Accomplishments that we're proud of

We're proud of building a fully AI-integrated application hosted on AWS Amplify, with Firebase for structured student and resource data. Our chatbot, powered by OpenAI’s API, delivers targeted recommendations, and we successfully designed a complete Figma UI while achieving our MVP. Despite AWS integration challenges, we fine-tuned AI responses and compiled an extensive resource folder from Brown’s Entrepreneurship Department, laying the groundwork for future expansion!

What we learned

[Florencia] I learned about AWS set up permissions and configurations, tools available such as BedRock Lambda and Lex, and how to host a Full-Stack dynamic app via Amplify. Also, learned a lot from the great work ethic of my teammates! :) [Earth] - I learned a lot of react and javascript today, especially how to interact with different APIs and databases. I also learned new ways to prompt LLMs and which parameters I should use for which tasks (ie. temperature and top_p).

[Nina] I learned a lot about different options for databases we can use for web apps (and how hard they can be to setup)!

What's next for GuidEd

Next steps for GuidED include building a secure platform, piloting the prototype in universities, and developing a mobile app. We’ll conduct user testing to refine features, enhance AI-driven recommendations, and improve the overall experience. After securing a partnership with Brown, we plan to scale rapidly and expand to more institutions.

Share this project:

Updates