Generative AI Use Cases with pgvector, Aurora PostgreSQL and Amazon Bedrock

Explore powerful Generative AI applications using pgvector on Amazon Aurora PostgreSQL with Amazon Bedrock

🌟 Overview

This repository demonstrates sample code implementations using pgvector, a powerful open-source PostgreSQL extension for vector similarity search. pgvector seamlessly integrates with PostgreSQL's native features, enabling sophisticated vector operations, indexing, and querying capabilities.

📚 Resources

• 📖 AWS Blog Post: Leverage pgvector and Amazon Aurora PostgreSQL for NLP, Chatbots and Sentiment Analysis
• 🎓 AWS Workshop: Generative AI Use Cases with Aurora PostgreSQL and pgvector

🚀 Use Cases

This repository showcases the following sample code implementations:

1. Product Recommendations 🛒

   • Implement intelligent product recommendation systems
   • Leverage vector similarity for personalized suggestions

2. Retrieval Augmented Generation (RAG) 🔄

   • Enhance LLM responses with relevant context
   • Implement efficient vector-based information retrieval

3. Semantic Search and Sentiment Analysis 🧠

   • Deploy sophisticated natural language search capabilities
   • Perform nuanced sentiment analysis on text data

4. Knowledge Bases for Amazon Bedrock 📚

   • Build scalable knowledge management systems
   • Integrate with Amazon Bedrock for enhanced AI capabilities

5. Movie Recommendations 🎬

   • Implement ML-based movie recommendation systems
   • Combine Aurora ML with Amazon Bedrock for sophisticated predictions

6. Democratizing Data Insights with Amazon Q Business 💼

   • Connect Amazon Q Business with Aurora PostgreSQL for enterprise-wide data access
   • Implement secure data exploration through user management and access control lists (ACLs)

🛠️ Getting Started

1. Clone the repository:

git clone https://github.com/aws-samples/aurora-postgresql-pgvector.git
cd aurora-postgresql-pgvector

2. Follow the setup instructions in each use case directory for specific implementation details.

🤝 Contributing

This repository is maintained for educational purposes and does not accept external contributions. However, you are encouraged to:

• Fork the repository
• Adapt the code for your specific needs
• Share your learnings with the community

📄 License

This project is licensed under the MIT-0 License - see the LICENSE file for details.

🔗 Related Projects

• pgvector
• Amazon Aurora
• Amazon Bedrock

---

Note: This repository is provided as-is and is intended for educational and demonstration purposes.