AI Builder Lab - Building AI Agents with Pulumi and AWS

Register on the AICamp Event Website is REQUIRED for admission.

Most AI agent demos look great in a notebook, but getting them into production means dealing with authentication, secrets, IAM policies, container builds, observability, and repeatable deployments. The proof of concept rarely covers any of that.

Join us for an evening of practical content on getting AI agents from prototype to production. Expect a hands-on workshop where you'll deploy a real multi-agent system on Amazon Bedrock AgentCore, with all infrastructure defined as code using Pulumi.

Agenda:
- 5:30pm~6:00pm: Checkin, food and networking
- 6:00pm~6:15pm: Welcome + intro
- 6:15pm~7:30pm: Hands-on workshop building AI Agents
- 7:30pm~8:30pm: Open discussion, Mixer and Closing.

What You'll Learn:

• Go from a minimal agent to a multi-tool, multi-agent system on Amazon Bedrock AgentCore
• Define your AI infrastructure as code, and why it matters for agent workloads
• Patterns for authentication, secrets, IAM, and observability when running agents in production
• Structure agent workflows that coordinate tasks, call external tools, execute code, and maintain state over time

Who Should Attend:
Engineers, platform teams, DevOps practitioners, and AI/ML developers who want to move past notebook demos and deploy real AI agents to production. Useful for anyone building on AWS, evaluating Bedrock AgentCore, or bringing infrastructure-as-code discipline to an AI workload.

Hands-on Workshop: Deploying AI Agents on AWS with Pulumi and Amazon Bedrock AgentCore
Instructor: Laci Videmsky — Solutions Architect, Pulumi Abstract: In this hands-on session, you'll go from a minimal agent to a multi-tool, multi-agent system running on Amazon Bedrock AgentCore, with all infrastructure defined as code using Pulumi. You'll build and deploy agents that coordinate tasks, call external tools, execute code, and maintain state over time.
You'll walk away with:
- A working multi-agent system deployed on AWS
- Infrastructure as code you can reuse and extend
- Hands-on experience with Amazon Bedrock AgentCore and Pulumi