Join us for Beyond resolution: Building the Self-Improving AI Agent

Beyond resolution: Building the Self-Improving AI Agent Wednesday 23 September, 18:30 CEST — Amsterdam Most CX teams have already shipped an AI agent that answers questions. That part is no longer hard. What separates the teams pulling ahead is what happens in week three, and week thirty — whether the agent quietly gets better on its own, or whether every improvement still sits in a queue behind someone's roadmap. This dinner starts at that beyond resolution: building the self-improving AI agent that learns from every conversation, finds its own gaps, and safely expands what it can resolve without waiting to be told. Getting there turns out to be less of a model problem and more of an operating one. Who owns the agent's performance. What coaching it looks like when the thing you're coaching isn't a person. What breaks in the org chart when improvement stops being a quarterly project and becomes something that happens continuously. Mike Murchison, CEO of Ada, moderates. The setting is closed and off the record, which usually means you get the version people share over dinner rather than the version they'd put on a slide. Held during HumanX Amsterdam, so many of you will already be in the city. Seats are limited to keep the conversation focused. Around the TableCX and digital leaders who have moved past the pilot stage and are now living with the operational reality of it — the people accountable for resolution rates, for the support headcount conversation, and for explaining to the board why an AI agent that works today should work better next quarter. Expect a mix of customer experience, digital transformation, product and technology executives, most of them wrestling with the same handover between automation and judgement. If any of that sounds like your last three months, you'll be in good company. How the Evening Runs 18:30 CEST — Welcome Reception & Arrival 19:00 CEST — Seated Dinner & Moderated Executive Discussion About the HostAda builds AI agents for customer service, and works with enough enterprises through the same transition to have a genuinely useful vantage point — less speculation about where agentic CX is heading, more pattern recognition from teams who have already tried it. Mike has spent the better part of a decade on the question of what makes an automated conversation actually resolve something, which makes for a good person to have steering the table.