SnowBrain is an open-source prototype designed to transform the way you interact with databases by making SQL queries conversational. It utilizes modern technologies such as Snowflake, OpenAI, and FastAPI to provide a seamless experience in data analysis. This guide will walk you through the process of getting SnowBrain running on your local machine.

Understanding the Tech Stack

Before diving into the installation steps, let’s quickly break down the tech stack that powers SnowBrain:

• Snowflake: For data storage and cloud management.
• Next.js: Serving a robust frontend and backend.
• Supabase: The database solution for persisting chat messages.
• Tailwind CSS: For sleek styling of the application.
• Pinecone: The vector database that enhances search capabilities.
• OpenAI: Utilized for advanced language processing.
• Langchain: Acts as a wrapper for the language model.
• Cloudinary: Manages image data efficiently.
• Clerk.dev: Handles secure user authentication.
• Upstash Redis: Manages rate limiting for API requests.
• FastAPI: A high-performance web framework for building APIs.
• Modal Labs: Hosts the backend FastAPI.
• Vercel: For easy deployment of your application.
• Umami: Provides web analytics for tracking user engagement.

Installation Steps

Ready to set up SnowBrain? Follow these steps carefully to get your personal data analyst working:

1. Update Environment Variables
   Make sure to update the environment variables as necessary. Refer to the provided example:

   bash
       .env.example

2. Auto Fetch All Schema DDL
   Run the following command:

   bash
       python3 embed/snowflake_ddl_fetcher.py

   Ensure you’ve installed the required dependencies:

   bash
       pip3 install -r embed/requirements.txt

3. Convert DDL Documents to Vector and Upload to Pinecone
   Use the following command:

   bash
       python3 embed/embed.py

4. Install Dependencies for the Code Plugin
   Navigate to the code-plugin directory and install necessary dependencies using Poetry:

   bash
       cd code-plugin
       poetry install

5. Deploy FastAPI to Modal Labs
   Deploy your FastAPI application with:

   bash
       modal deploy main.py

   Make sure to add a secrets file in Modal Labs and store your endpoint in environment variables:

   bash
       MODAL_API_ENDPOINT=
       MODAL_AUTH_TOKEN=random_secret

6. Install Packages
   Run the following command to install packages:

   bash
       bun install

7. Run Locally
   Test your setup with:

   bash
       bun run dev

   Then test the build:

   bash
       bun run build

8. Deploy to Vercel
   When you’re ready, deploy the project to Vercel. Note that Vercel build is automatically blocked on certain folders and files. Consider adding a build block command in the Vercel dashboard.

Example Queries

SnowBrain is designed to simplify SQL querying. Here are a few examples you can try:

• Total revenue per product category: Show me the total revenue for each product category.
• Top customers by sales: Who are the top 10 customers by sales?
• Average order value per region: What is the average order value for each region?
• Order volume: How many orders were placed last week?
• Product price listing: Display the list of products with their prices.

Troubleshooting Tips

If you encounter issues during installation or running SnowBrain, consider these troubleshooting tips:

• Ensure all environment variables are correctly set.
• Check that all dependencies are installed properly using the relevant package managers.
• If the API doesn’t respond, verify your Modal Labs configuration.
• Refer to your terminal output for error messages, which often provide clues for resolutions.
• For any persistent issues, check the official issues page on GitHub for help.
• For more insights, updates, or to collaborate on AI development projects, stay connected with fxis.ai.

Conclusion

Now that you have SnowBrain set up, you can start enjoying a new way of querying and analyzing your data! This innovative tool not only makes data exploration easier but also enhances your ability to interact with your dataset through a conversational interface.

At fxis.ai, we believe that such advancements are crucial for the future of AI, as they enable more comprehensive and effective solutions. Our team is continually exploring new methodologies to push the envelope in artificial intelligence, ensuring that our clients benefit from the latest technological innovations.