Building Production-Ready AI Agents
summary
Strands is an open source framework for production agents that runs on Python or TypeScript and is model-agnostic. Live coding shows an agent spun up with a single line of code, tools added by decorating a function with @tool decorator, memory and session management handled by the SDK, built-in observability and MCP client, multi‑agent patterns and agents create agents demos, and an option to deploy to a managed cloud runtime (Agent Core) that actually runs and scales the agent.
AI-written summary · about
chapters (11)
- 0:00Introduction to agentic AI
- 1:14Defining agentic AI systems
- 2:34When to use agentic AI
- 3:55Strands SDK overview
- 7:18Core features of Strands
- 10:12Creating custom tools
- 12:29Self-extending agents
- 15:14Multi-agent orchestration
- 21:37Memory and MCP support
- 22:35Case study: Marginalia tool
- 25:51Production deployment
related talks
14mIntroducing Strands Agents, an Open Source AI Agents SDK
16mBuilding Agentic AI applications with Positron and AWS Strands Agents
19mBuilding Agents at Cloud Scale
21mServerless Agents: Real-World Tooling with Strands SDK, MCP & AWS
06mKeynote: Building Agents at Scale
22mContext Aware Development in Kiro
23mBuilding AI Agents in TypeScript
46mFullstack AI: The Building Blocks for Production Ready AI-Powered Software
