Agentic Apps: How Big Companies Are Actually Deploying AI Agents

Pretty much every AI agent I've built for myself is one agent I chat with. That's great for my own work, but it's not how companies actually adopt AI a lot of the time. So in this video I go through the lessons I end up teaching every company I work with, and how that takes shape as what Oracle calls agentic apps! Multiple agents working toward one goal, AI only where you need judgment, rules turned into actual code with tests, and a person approving anything that matters. ~~~~~~~~~~~~~~~~~~~~~~~~~~ - Oracle AI for Fusion Applications: https://social.ora.cl/6002BGfjvg Thanks to Oracle for working with me on this! ~~~~~~~~~~~~~~~~~~~~~~~~~~ Join Dynamous, my AI community for mastering agentic engineering and building AI agents: https://dynamous.ai ~~~~~~~~~~~~~~~~~~~~~~~~~~ 0:00 Why One Chat Agent Isn't Enough 0:45 What Is an Agentic App? 2:47 Building an App in Oracle AI Agent Studio 3:27 Inside the Agentic App 4:36 Agents That Prioritize Your Next Actions 5:14 Every LLM Call Runs Inside a Workflow 7:17 Security with Policies 8:20 Policy Nodes vs. RAG 8:57 Building the Leave Policy Workflow 9:30 Workflow Permissions and Triggers 10:22 Visual Outputs, Not Walls of Text 11:40 Recap and What's Next ~~~~~~~~~~~~~~~~~~~~~~~~~~ Join me as I push the limits of what is possible with AI. I'll be uploading videos weekly - at least every Wednesday at 7:00 PM CDT!