488. Agentic AI Organization Design

In this episode, host Michael Marchuk speaks with Tim Shea, founder and CEO of Latticework Insights, about what it takes to build an enterprise-grade agentic AI program that delivers measurable value. Shea explains that while LLMs are powerful, hallucinations, non-determinism, and high-stakes risk make many agentic initiatives fail without human checks, clear leadership, and strong operating models. He outlines how to decide when a use case truly needs agentic AI versus deterministic automation, using an analytics workflow (extraction, modeling, interpretation, evaluation) to show where agents help most. Shea shares a restaurant-chain example where an agent rapidly replicated unit tests across pipelines to improve data quality and executive confidence. He warns against rushing, token-burning for its own sake, and losing focus on business outcomes, emphasizing upskilling and domain expertise as key differentiators over the next 12–18 months. What we talked to Tim about: -What Agentic AI Means -Why AI Fails in Production -When You Need Agents -Agent Roles in Analytics -Managing Agent Swarms -Org Design and Upskilling -Measuring Real Outcomes -Common Adoption Mistakes -Workflow Shifts and Tools -High Stakes Industries Limits Visit us on our socials: 🦾 Get started with SS&C Blue Prism: https://okt.to/JcMLdU 🧑‍💻LinkedIn: https://okt.to/k8zIdp ✖️Twitter: https://okt.to/fHyd9G 🙋‍♀️Facebook: https://okt.to/Vyjfiz 📸Instagram: https://okt.to/5nYvIf 💭Blog: https://okt.to/QuGqVP 🤩Case studies: https://okt.to/ft1AMX To ensure that you never miss an episode of Transform NOW, be sure to subscribe!