Behind Chat Syllabus: Building a Smart Course Assistant

Overview

At the start of every semester, students face the challenge of managing multiple courses and their corresponding syllabi. Why not leverage the power of generative AI to simplify this process? Enter Chat Syllabus—a tool designed to analyze your syllabus and provide actionable insights, such as course difficulty, weekly assignments, and workload. Say goodbye to uncertainty and let Chat Syllabus help you plan smarter, stay organized, and tackle your semester with confidence.

Features

Course Analyzer: Assess course difficulty on a scale of 1-10, where 10 represents the most challenging. Plan your semester accordingly!
Contact Info: Easily access the email addresses and phone numbers of your instructors, professors, and TAs. Get help when you need it.
Grade Breakdown: Gain a clear understanding of how each component contributes to your final grade. Focus on what matters most.
Assignments To-Do: Stay on top of your weekly assignments with a detailed checklist. Never miss a deadline again!
Chatbot: Have questions? Ask anything to the chatbot for instant answers.
And More: Explore additional features designed to make your academic life easier.

Technologies

Chat Syllabus is a web application powered by advanced AI models. Here's an overview of the technologies we use to deliver a seamless experience:

Frontend

Our frontend is built with React, enabling dynamic rendering of web pages. We deploy the application using Vercel and Cloudflare for optimal performance. Additionally, the frontend handles the storage of text embeddings for uploaded syllabi directly in the user's browser. This eliminates the need for a backend database, ensuring data privacy and efficiency.

Backend

The backend is powered by Python and the FastAPI framework, managing requests from the frontend. As a Retrieval-Augmented Generation (RAG) application, we utilize OpenAI's Davinci model for text embeddings and deploy Groq's LLaMa 3.3 8B and LLaMa 3.3 70B models for fast, accurate text-based responses.

Our Team

Haozhe Li: Architect & Full-Stack Developer
Hansen Punnoose: Chatbot & Frontend Specialist
Zhicheng Hu: RAG & Course Analyzer Expert
Steven Zheng: Grade Breakdown & Assignment To-Do Lead