Environment setup

Let's set up your development environment. We'll clone the repository, create a virtual environment, and configure your API keys.

Dev environment stack

The four layers of the development environment

terminal
bash
# Clone the workshop repository
git clone https://github.com/learnwithparam/text-to-sql-agentic-rag-chatbot.git
cd text-to-sql-agentic-rag-chatbot

# Copy environment template and add your API key
cp .env.example .env
# Edit .env with a free key from https://aistudio.google.com/apikey

# Install dependencies
make install

# Download dataset and initialize database
make setup-data

Clone the repository, configure your API key, and set up the project with uv.

With the repo cloned and dependencies installed, the last step is configuring your API key. The .env file tells LiteLLM which model provider to use.

.env
bash
# Get a free API key from https://aistudio.google.com/apikey
GOOGLE_API_KEY=your_key_here

# Default model (Google Gemini has a generous free tier)
DEFAULT_MODEL=gemini/gemini-2.5-flash

Edit your .env file with an API key. We recommend Google Gemini for its free tier, perfect for learning.

We use LiteLLM to abstract the LLM provider. This means you can switch between OpenAI, Anthropic, Google, and other providers just by changing the DEFAULT_MODEL in your .env file!

Validation checklist: Environment setup checklist

Loading practice…

Your development environment is ready. Everything is set up for building the Agentic RAG system.

Validation checklist: Introduction checklist

Loading practice…