Building Agents at Cloud Scale
summary
Demo shows a code-assistant agent shipped to CLI in 3 weeks, using an open-source Strand agents SDK that defaults to Amazon Bedrock but also plugs into OpenAI and Llama. The SDK includes 20+ prebuilt tools (memory, multimodal, retrieve) and a pattern where a knowledge base lets an agent pick relevant tools from 6,000 descriptions. Production pattern uses an MCP Lambda handler behind API Gateway; demo invokes a remote roll_d20 tool and returns a 7, proving the pipeline works end to end.
AI-written summary · about
chapters (9)
- 0:00AI agent innovation
- 1:10Reimagining Alexa
- 3:26Scale and expert systems
- 5:15Amazon Q developer
- 7:20Introducing strands agents
- 9:19Strands agent features
- 11:59Model context protocol
- 14:22Serverless agent tools
- 17:23Future of interoperability
related talks
21mServerless Agents: Real-World Tooling with Strands SDK, MCP & AWS
15mAgents Don’t Ship Themselves
16mBuilding Agentic AI applications with Positron and AWS Strands Agents
06mKeynote: Building Agents at Scale
28mBuilding Agentic AI applications on AWS
40mAgents in Action: Advancing Open Source AI Missions
29mBuilding Production-Ready AI Agents
38mBuilding AI Agents and Assistants for DevOps Workflows