Control the Conversation: Using an Agentic Approach to Talk Complex Data with AI Chatbots


Sure, you’ve all tried asking them a few basic questions for fun, but have you ever wondered what it would be like if you could truly talk to those new LLM-powered chatbots about your data and get accurate answers back, consistently and securely? Now’s your chance to learn how! Join Marcus Wilkins, InformData’s senior business intelligence engineer, as he discusses how to build a workflow that uses multiple AI agents alongside conventional coding to build the perfect response directly from your data—all without ever sharing it with any external sources! Marcus will discuss the theory behind this approach, then follow with a live demonstration using InformData’s actual Domo environment.
Here are the key takeaways:
- Why an agentic workflow is better than a traditional interaction with a chatbot.
- What are some key design principles and concepts when building an agentic workflow.
- How to ensure the responses your AI agents provide are consistent and trustworthy.


Marcus Wilkins works as InformData’s lead data scientist and has spent a good number of years in various data science and engineering roles across the realms of robotic process automation, software development, business analysis, and background investigation. His current focus is the development and integration of AI-powered programs into InformData’s Domo ecosystem; he also works closely with key company leadership to deliver accurate and actionable business intelligence data to the organization. Marcus lives on the East Coast of the United States and loves a good cup of coffee, a good book, and the Great Outdoors.
Domo transforms the way these companies manage business.