This folder includes resources and examples to assist organizations in transitioning Amazon Bedrock-powered applications from proof-of-concept (PoC) to production-ready solutions. It offers tools, best practices, and sample code to guide the effective implementation, testing, validation, and operationalization of Bedrock applications.
-
Inference Profiles: This notebook provides a comprehensive guide to:
- Create and Configure Inference Profiles: Understand and apply profile options for optimal application performance.
- Invoke Models with Inference Profiles: Validate model invocation with real-time response streaming.
- Operationalize Tagging and Cleanup: Implement organized tagging for resource management and automate profile cleanup as needed.
-
Bedrock Batch Orchestrator: This CDK stack helps facilitate large-scale batch inference tasks by automating end-to-end processing with Step Functions and EventBridge.
We welcome community contributions! Please ensure that your sample aligns with AWS best practices. Update the Contents section of this README file with a link to your sample, along with a brief description.