🍴 Charlotte Eatz 🤖

Eric P., product mgr. | Yaxin Z., data/prompt engr. | Lakshmi J. K., app dev. | Gaurav S., LLM ops.

🎉 Final Report | 🚀 Streamlit App

Welcome to Charlotte Eatz, your go-to app for exploring Charlotte's tastiest spots! We aim to simplify dining experiences in Charlotte, NC by seamlessly integrating restaurant reservations with additional end-to-end services like transportation booking and retrieving restaurant reviews. The core value and heart of Charlotte Eatz lies in its ability to streamline multiple tasks through DineBot, the app’s chatbot, saving time and improving users’ foodie experience.

DineBot, the heart of Charlotte Eatz. Generated by DALL-E.

✨ Ready to Get Started?

1. Streamlit Cloud ☁️

The quickest and easiest way to our app is through Streamlit Cloud!

Simply input your OpenAI API Key and you are ready to chat with DineBot 🤖!

2. Local Deployment ⚙️

First, clone the repo like this:

git clone --depth 1 https://github.com/dsba6010-llm-applications/AgenticRAG-CharlotteEatz.git

Warning

Our virtual environment was accidentally included in the initial push. It has since been removed from the repo but will be present in git history. Be sure to include --depth 1 when cloning the repo to exclude git history and avoid downloading the virtual environment.

Then cd into the folder AgenticRAG-CharlotteEatz.
Create a virtual environment, activate it, and install the dependencies.

python3 -m venv venv
source venv/bin/activate
python -m pip install --upgrade pip
python -m pip install -r requirements.txt

Tip

If you're using Windows CMD, the 2nd line will be .\venv\Scripts\activate.bat. Alternatively, if you're using Windows PowerShell, it would be .\venv\Scripts\activate.ps1

Lastly, create a .env file with OPENAI_API_KEY='Your OpenAI API Key'.

You are now able to run our app locally using streamlit run dinebot_app.py!

🛠️ Behind the Scenes

• Core Architecture: Powered by an agentic RAG (Retrieval-Augmented Generation) system made by and for Charlotte foodies.
  • Built using FAISS for fast and efficient information retrieval.
  • Developed with LangChain for agents, tools, and overall advanced conversational flows.
  • Leveraging OpenAI's GPT model for easy integration and support.
  • Using Streamlit for a simple and accesible user experience.

For more details (or if you are just curious), please take a look at our Final Report!

---

💕 Thanks for using Charlotte Eatz! Let’s make every meal an adventure with DineBot. 🍷🍕