

Guardrails, Logging and Hallucinations: Running AI Agents Safely at Work
A live online lunchtime masterclass for leaders and teams. Join from anywhere, just watch and ask questions.
AI agents now take real actions inside companies. That's powerful, and risky. In one lunchtime you see how teams keep AI agents safe: guardrails that stop bad actions, logs you can audit, and ways to catch made-up answers before customers see them.
What you'll learn
Guardrails: filters, permissions and approvals for risky actions, shown live
Logging: what to record so you can explain any AI decision later
Hallucinations: why models make things up, and how to ground and test answers
Prompt injection: the attack every team should know, explained without jargon
Who it's for
Executives, team leads, and heads of IT, security, compliance and operations. No coding background needed.
Agenda (Berlin time)
13:00 - 14:00 · Masterclass: talk and live demo
14:00 - 14:30 · Q&A
Free entry
This one is free. Register to get the link.
Your host
Aemal Sayer has been building AI applications for 10 years and AI agents for the last 4, with enterprise teams since 2016. More on LinkedIn.
Go deeper, in Venice
AI for Executives: Venice Masterclass, Thu 8 April. An intensive evening on an island in the Venice lagoon, with dinner and wine. See the masterclass →