Your First AI Agent Watch It Think, Act, and Work

LatinaTechies AI Upskilling Series · Event 3 of 3 You've used AI as a smart assistant. Now meet AI that actually does things. Most people interact with AI by asking it questions. An agent is different — it uses tools, makes decisions, and completes multi-step tasks without you guiding every click. That shift from AI that answers to AI that acts is one of the most significant things happening in tech right now, and understanding it is becoming a real career differentiator. In this session, Mel Li — AI/ML Solutions Architect at AWS — walks through how enterprise agents actually work, using the same tools being used to build production agents today: OpenClaw for agent orchestration, Claude Code for AI-powered development, and Amazon Bedrock AgentCore for hosting and running agents at scale. What's covered: What makes something an AI agent — and how it's fundamentally different from a chatbot How OpenClaw, Claude Code, and Amazon Bedrock AgentCore work together to build and run a real agent How the same pattern that powers personal tools scales to enterprise automation How to talk about agentic AI confidently — in interviews, at work, and beyond Event Resources: LLM-Harness Paper: https://picrew.github.io/LLM-Harness/ Sample Agent GitHub: bit.ly/4e5Ztcd About LatinaTechies LatinaTechies is a volunteer-run nonprofit offering free virtual AI upskilling events for Latina professionals. Learn more and register for upcoming events at latinatechies.org.