• 

  Landing Page

• 

  Transfer Institution Selection

Assist AI: Your personal AI-powered Transfer Counselor

💡 Inspiration

As community college students, we know firsthand how frustrating and time-consuming it can be to find all the information you need to prepare for the college transfer process. Every single college has its own course articulations and specific requirements, which quickly becomes confusing and stressful as students try to plan their coursework and applications.

Simply put, the tools available to students aren't enough. From navigating conflicting counselor information to managing the overwhelming volume of forms and websites they have to manually cross-reference, the existing process is broken. We knew there had to be a better way, and so we built one.

🎯 What it Does

Assist AI is a powerful tool that aggregates transfer information across American universities and course articulation agreements from ASSIST.org to provide the first truly all-in-one solution for transfer students.

Key Features

🛠️ How We Built It

The frontend for our site was built using next.js, tailwind, and eslint, which we deployed using Vercel server hosting.

Data Pipelines

1. Course Articulations (Planner Data):

   • A python script is used to scrape course articulation information from assist.org, using the BeautifulSoup and HTML parser modules.
   • These are stored on a PostgreSQL database, which is formatted into the course planner on the frontend using python scripting.

2. General Transfer Info (Chatbot Data):

   • A separate python script uses the Gemini API to search through university sites and scrape relevant information for each university, which it stores as raw text.
   • This raw data is stored in a backend on Supabase, which is used to embed the raw text files into the OpenAI API for parsing.
   • This connects to the chatbot feature on the frontend, which answers users' transfer questions based on this data.

Built With

• beautiful-soup
• next.js
• openai
• postgresql
• python
• supabase
• vercel