Building AI Agents: No Code Required

AI is moving from chat to action. The same models that answer questions can now read an inbox, pull data from a spreadsheet, draft a response, file a document, and hand the result to a person for approval. Systems that do this are called agents, and they are rapidly becoming part of ordinary professional work beyond engineering teams.

This workshop teaches professionals to build working agents using no-code automation platforms and the AI tools their organizations already have. Across two hands-on sessions, you will build agents against your own recurring tasks: triaging messages, monitoring sources, assembling reports, and moving information between the systems you use every day.

The house emphasis on verification applies with particular force here: an agent that acts on your behalf needs scoping, oversight, and verification. You will learn how to build agents, and how to decide what they should and should not be trusted to do. You will leave with two to three working agents, a verification habit, and a one-page governance checklist for everything you automate next.

Participants will learn to:

• Distinguish an agent from a chatbot and from simple automation, and recognize which tasks suit each
• Build a working agent on a no-code platform: triggers, instructions, tools, and memory
• Connect agents safely to email, files, calendars, and spreadsheets using least-privilege access
• Design multi-step agent workflows with branching, error handling, and verification
• Place human-in-the-loop checkpoints where judgment, risk, or money is involved
• Estimate and govern running costs, and recognize when a task should not be automated at all
• Apply a governance checklist covering data scope, permissions, logging, and failure modes