You're a startup. You have a small team, limited runway, and customers who need value now, not after a six-month agent platform build. You need AI patterns that are fast to ship, cheap to run, and easy to iterate on without a dedicated ML team.

Join us for an action-packed afternoon starting with lunch and open networking, followed by hands-on building. We'll cover the full lifecycle of a production AI agent, from prompt to deployment using open-source tools (LangGraph, LlamaIndex) and AWS services like Amazon Bedrock, Lambda, and DynamoDB. Every lab is grounded in real startup use cases: customer-facing chatbots, internal copilots, automated workflows, and RAG over your own data.

We'll also be honest about the hard parts: when agents fail, how to eval without a data science team, and how to control costs before they spiral.

What you'll leave with:

  - A working AI agent deployed on AWS 

  - Production patterns for memory, tool use, evals, and observability

  - A clear picture of AWS costs 

  - 1:1 time with an AWS solutions architects

  - Connections to other builders/founders solving the same problems