Announcing MCP Toolbox Java SDK v1.0: Agentic data access for the enterprise
cloud.google.com·6d ago
Cloud & Infrastructure
AWS Blog·Aug 6·5 min read
AWS Blog·Aug 6·5 min readRuntime instances: persistent compute for production AI agents on Amazon Bedrock AgentCore
TL;DR
Amazon Bedrock AgentCore introduces runtime instances, a managed compute option that allows AI agents to persist state and collaborate over extended periods, enhancing production workflows.
✦ Why It Matters
Start using runtime instances today to simplify your AI agent deployment and management.
Key Takeaways
- Utilize runtime instances for persistent AI agent workflows.
- Leverage GPU acceleration for compute-intensive tasks.
- Deploy multiple agents on a shared infrastructure easily.
- Integrate with existing AgentCore APIs for seamless orchestration.
How It Works
Runtime instances provide AWS-managed EC2 infrastructure where multiple agents can run in a single environment. Each agent operates with its own dependencies and can share a file system for collaboration.
This setup allows agents to persist state across sessions, enabling complex workflows that can last up to 14 days. Developers can package agents with minimal effort, using a simple decorator and a zip file or container image.