SmartyPants

Build and manage AI solutions fast without complicated learning curve about AI. An open source initiative by Shaharia Lab OÜ

shaharia-lab/smarty-pants

What is SmartyPants?

SmartyPants AI is an intelligent, AI-driven platform that seamlessly integrates multiple data sources, embedding models, and LLM providers. It offers powerful semantic search capabilities and an intuitive chat interface, allowing users to easily configure and interact with various AI models. Whether you're building a smart chatbot or need advanced data processing and querying, SmartyPants provides a flexible, user-friendly solution for your AI-powered applications.

Why SmartyPants?

We named this project SmartyPants. A lighthearted name implying the system is incredibly intelligent, able to handle complex queries with ease.

Key Features:

• Multi-source data integration and embedding generation
• Configurable LLM and embedding models
• Semantic search functionality
• Built-in chat interface and chatbot creation tools
• Easy-to-use API for seamless integration

Empower your projects with SmartyPants – where AI meets simplicity!

Installation

• Pre-requisites:
  • PostgreSQL 13 or higher
    with pgvector extension enabled
    for vector search capabilities.

Use as a Docker Image

To use this application with Docker, follow these steps:

Pulling the Docker Images

1. For the backend:

   docker pull ghcr.io/shaharia-lab/smarty-pants-backend:$VERSION
   

2. For the frontend:

   docker pull ghcr.io/shaharia-lab/smarty-pants-frontend:$VERSION
   

Replace $VERSION with the desired version tag. All available versions can be
found here.

Running the Backend

To run the backend, you need to set the following environment variables:

• DB_HOST: Database host
• DB_PORT: Database port
• DB_USER: Database user
• DB_PASS: Database password
• DB_NAME: Database name

Run the backend with this command:

docker run \
  --name smarty-pants-backend \
  -p 8080:8080 \
  -e DB_HOST=<value> \
  -e DB_PORT=<value> \
  -e DB_USER=<value> \
  -e DB_PASS=<value> \
  -e DB_NAME=<value> \
  -e DB_MIGRATION_PATH=<value> \
  ghcr.io/shaharia-lab/smarty-pants-backend:$VERSION start

Replace <value> with the appropriate values for your environment.

Running the Frontend

To run the frontend, use this command:

docker run \
  --name smarty-pants-frontend \
  -e NEXT_PUBLIC_API_BASE_URL="http://localhost:8080" \
  -p 3000:3000 \
  ghcr.io/shaharia-lab/smarty-pants-frontend:$VERSION

Make sure to configure any necessary network settings to allow the frontend to communicate with the backend.

Environment Variables

📋 Development

Prerequisites

Backend

• Go 1.22 or higher
• PostgreSQL 13 or higher with pgvector
  extension enabled

Frontend

• Node.js 20 or higher

Running Locally

• Clone the repository and navigate to the project directory.

  git clone git@github.com:shaharia-lab/smarty-pants.git
  cd smarty-pants

• Start a compatible PostgreSQL database with the pgvector extension enabled. We have included a docker-compose.yml
  file in the root directory to help you set up the database. Run the following command to start the database:

  docker-compose -f docker-compose.yml up -d

• Create a .env file in the root directory and add the required environment variables. Load the environment variables
  using the following command:

  export $(grep -v '^#' .env | xargs)

• Run the backend:

  go run . start

If you want to run the frontend, follow these steps:

• Install the dependencies:

  cd frontend
  npm install

• Start the frontend:

  npm run dev

🚀 Observability

To enable observability, set the TRACING_ENABLED environment variable to true. This will enable tracing. You can also
enable metrics by setting the OTEL_METRICS_ENABLED environment variable to true. The metrics will be exposed on the OTEL_METRICS_EXPOSED_PORT port.

You can run any OpenTelemetry Protocol (OTLP) compatible tools to visualize the trace. We recommend using Jaeger.

docker run -d --name jaeger \                   
  -e COLLECTOR_OTLP_ENABLED=true \
  -p 16686:16686 \
  -p 4317:4317 \
  -p 4318:4318 \
  jaegertracing/all-in-one:latest

🤝 Contributing

Contributions are welcome! Please follow the guidelines outlined in the CONTRIBUTING file.

📝 License

This project is licensed under the MIT License - see the LICENSE file for details.