As agentic AI systems move from research to production, engineering teams need the skills to design, control, and scale them safely. This hands-on session will guide you through building an autonomous AI agent for a realistic enterprise use case from architecture design to live deployment. By the end, you’ll have built and tested a fully functioning agent that can plan, call tools, and execute tasks with guardrails in place. You’ll learn how to: → Design an end-to-end architecture using MCP or UTCP for seamless tool integration. → Implement planning, memory, and execution layers for real-world workflows. → Add governance and fail-safe mechanisms to prevent unintended actions. → Deploy your agent in a sandbox environment and evaluate performance in real time.