AztecPlanner

An intelligent academic planning assistant built for SDSU students, powered by AI and modern web technologies.

Created for ACM SDSU Innovate 4 Hackathon 2025, AztecPlanner revolutionizes course planning through AI-driven recommendations and intelligent degree audit analysis.

Architecture

AztecPlanner features a cutting-edge full-stack architecture leveraging the latest in web and AI technologies:

Frontend: React + Vite

Lightning-fast React application built with Vite for instant hot-module replacement and optimized production builds. Handles dynamic user interactions, degree audit parsing, and real-time semester planning with a responsive, intuitive interface. Styled with Tailwind CSS and featuring interactive data visualizations powered by Recharts.

Backend: FastAPI + Uvicorn

High-performance async Python webserver powered by FastAPI and Uvicorn ASGI server, featuring:

• Vector Database: ChromaDB integration for semantic course search and intelligent recommendations
• LLM Integration: LiteLLM endpoints enabling multi-model AI assistance
• Custom RAG Pipeline: Retrieval-Augmented Generation system combining vector search with language models for context-aware academic advising

Project Structure

AztecPlanner
├── front-end-webserver/     # React + Vite frontend application
│   ├── Smart degree audit parsing and requirement tracking
│   ├── Interactive semester planning interface
│   ├── AI-powered course recommendations
│   ├── Real-time course search and filtering
│   └── Tailwind + Recharts for data visualizations and responsive design
│
├── utils/                    # FastAPI backend server
│   ├── ChromaDB vector store for semantic search
│   ├── Course database management
│   ├── LLM integration layer with LiteLLM
│   └── Custom RAG implementation
│
├── scraper/                  # Data acquisition pipeline
│   ├── Beautiful Soup web scraper for SDSU course catalog
│   └── Selenium-powered RateMyProfessor data extraction
│
└── tests/                    # API testing suite
    └── Comprehensive endpoint validation notebooks

Tech Stack

Frontend

• React 18 - Modern UI component library
• Vite - Next-generation frontend tooling
• Tailwind CSS - Utility-first CSS framework for rapid UI development
• Recharts - Composable charting library for data visualization

Backend

• Python 3.10 - Core backend language
• FastAPI - Modern, fast web framework
• Uvicorn - Lightning-fast ASGI server
• ChromaDB - AI-native vector database
• LiteLLM - Unified LLM API interface

Data Collection

• Beautiful Soup - HTML parsing and web scraping
• Selenium - Dynamic content extraction

Getting Started

Prerequisites

• Anaconda3 or Miniconda
• Node.js (v16 or higher)
• .env file with required API keys

Installation

1. Set up Python environment

conda create -n aztecPlanner python=3.10
conda activate aztecPlanner
pip install -r requirements.txt

2. Launch the backend server

python -m py_compile utils/api_server.py
python -m uvicorn utils.api_server:app --reload --host 127.0.0.1 --port 8000

3. Start the frontend development server (in a separate terminal)

cd front-end-webserver
npm install
npm run dev

Done! Access the application at http://localhost:5173

Features

• Intelligent Degree Audit Parsing - Automatically extracts and analyzes degree requirements
• Smart Semester Planning - AI-assisted course scheduling with prerequisite validation
• Semantic Course Search - Vector-based search powered by ChromaDB
• AI Academic Advisor - RAG-powered chatbot for personalized course recommendations
• Professor Insights - Integrated RateMyProfessor data for informed decision-making

---

Built with love for SDSU students by the Sudo Apt-get Win Team during the Innovate 4 SDSU Hackathon