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

Variable	Required	Default value	Description
APP_NAME	No	smarty-pants-ai	Name of the application
DB_HOST	Yes	"localhost"	Database host address
DB_PORT	Yes	5432	Database port number
DB_USER	Yes	"app"	Database user name
DB_PASS	Yes	"pass"	Database password
DB_NAME	Yes	"app"	Database name
API_PORT	No	8080	Port number for the API server
API_SERVER_READ_TIMEOUT_IN_SECS	No	10	API server read timeout in seconds
API_SERVER_WRITE_TIMEOUT_IN_SECS	No	30	API server write timeout in seconds
API_SERVER_IDLE_TIMEOUT_IN_SECS	No	120	API server idle timeout in seconds
TRACING_ENABLED	No	false	Enable or disable tracing
OTLP_TRACER_HOST	No	"localhost"	OpenTelemetry Protocol (OTLP) tracer host
OTLP_TRACER_PORT	No	4317	OTLP tracer port
OTEL_METRICS_ENABLED	No	false	Enable or disable OpenTelemetry metrics
OTEL_METRICS_EXPOSED_PORT	No	2223	Port to expose OpenTelemetry metrics
COLLECTOR_WORKER_COUNT	No	1	Number of collector workers
GRACEFUL_SHUTDOWN_TIMEOUT_IN_SECS	No	30	Graceful shutdown timeout in seconds
PROCESSOR_WORKER_COUNT	No	1	Number of processor workers
PROCESSOR_BATCH_SIZE	No	2	Batch size for the processor
PROCESSOR_INTERVAL_IN_SECS	No	10	Processor interval in seconds
PROCESSOR_RETRY_ATTEMPTS	No	3	Number of processor retry attempts
PROCESSOR_RETRY_DELAY_IN_SECS	No	5	Delay between processor retry attempts in seconds
PROCESSOR_SHUTDOWN_TIMEOUT_IN_SECS	No	10	Processor shutdown timeout in seconds
PROCESSOR_REFRESH_INTERVAL_IN_SECS	No	60	Processor refresh interval in seconds
ENABLE_AUTH	No	false	Enable authentication system for the application
GOOGLE_OAUTH_CLIENT_ID	No	""	Google OAuth client ID
GOOGLE_OAUTH_CLIENT_SECRET	No	""	Google OAuth client secret
GOOGLE_OAUTH_REDIRECT_URL	No	""	Google OAuth redirect URL
SUPER_ADMIN_EMAIL	Yes	""	Email address of super admin

📋 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.