What's genuinely shipping, and how to secure autonomy at scale
Every vendor seems to be branding their product as "agentic" these days, yet the real shift in 2026 is far more specific. A new generation of always-on, persistent, multi-channel autonomous assistants has quietly arrived, and it's already shaping enterprise conversations whether IT teams planned for it or not.
This GenAI Tech Meetup is dedicated to cutting through the noise and focusing on the questions that genuinely matter. What does harness engineering look like in practice, and where is it delivering real value? What is OpenClaw, and how do the tools emerging in its wake truly compare? And with a deployment landscape crowded by chatbots, no-code orchestrators, and Agent Developer Kits, which of these are actually moving the needle in production?
The past year has reshaped the field at remarkable speed, and as these systems grow more capable, the surfaces they expose grow with them. This edition takes a practical look at what automated security testing for agentic systems really involves, what is genuinely shipping today, and the hard-won lessons from building security into autonomous pipelines at scale.
Is this meetup for you?
This meetup is built for AI and ML leads, platform and ML Ops engineers, heads of data, CTOs, and product leaders responsible for bringing GenAI into production inside complex or regulated organisations. It's the right room for those technical enough to care about how the sausage is made, and senior enough to decide whether to build it or buy it.
What you will take away
- A behind-the-scenes look at what ADC is doing with agentic AI today, and the impact it's already creating.
- An in depth look at how to deploy agents securely at scale, from our expert friends at orq.ai
- Meaningful peer conversations with teams navigating the same deployment challenges, continued over a relaxed walking dinner.
Please note that this is an in-person event.
Join ADC's Agentic AI Tech Meetup
Seats are limited. Reserve yours for an evening of real insight and good conversation.
From Orq.ai and ADC
Hear directly from the industry experts.
Edward Jansen, Technology & Innovation Lead | ADC
Edward Jansen is Lead of Technology & Innovation at ADC Consulting in Amsterdam, where he leads the firm's GenAI and agentic strategy work for enterprise clients. He co-authored ADC's Data & AI Strategy Framework and writes regularly on the realities of AI adoption in professional services. He is a frequent speaker on enterprise AI strategy.
Edward's LinkedIn
Bauke Brenninkmeijer, AI Research Engineer | Orq.ai
Bauke Brenninkmeijer is an AI Research Engineer at orq.ai in Amsterdam, where he leads the development of the company's open-source red teaming framework for agentic systems. He brings six years of experience across NLP and MLOps, with previous roles at ABN AMRO, Sytac, ING, and his own startup OneTwoModel, along with an academic background in Data Science from Radboud University. Bauke is an active contributor to the MLOps Community Amsterdam and a regular speaker on agentic AI, evaluation, and security.
Bauke's LinkedInAbstract
Agentic AI in 2026: the state of play
Two years into the agentic wave, the noise has settled enough to see what's actually working. Coding harnesses are in every developer's terminal. Managed agents from the hyperscalers are landing in enterprise pilots. A new category of always-on autonomous agents is quietly emerging. Edward maps the 2026 landscape: the categories that matter, where production deployments are real, where the gap between demo and value is still wide, and what enterprises should be buying versus building. Expect a clear-eyed view from inside client engagements at major Dutch and European institutions.
Red teaming your AI agent before it ships
When your LLM has tools, the threat model changes. The question is no longer whether you can trick the agent into saying something problematic, but whether you can trick it into doing something harmful. A chatbot that leaks its system prompt is an embarrassment, yet an agent that forwards customer emails to an attacker or approves fraudulent payments is a business-ending incident.
In this session, Bauke walks through automated security testing for agentic systems, covering the OWASP LLM Top 10, the new OWASP Agentic AI Top 10, and the attack categories that genuinely matter once an LLM can act, including goal hijacking, tool misuse, memory poisoning, and cascading failures. He'll also share how to stand up a red teaming pipeline before your agent ships, with multi-turn attack walkthroughs from the open-source framework his team has built at orq.ai, alongside practical tooling you can put to work straight away.
Programme
17:00 – 17:30 | Arrival and welcome, with time to explore the Demo Marketplace
17:40 – 17:50 | Introductory remarks
17:50 – 18:30 | "Agentic AI in 2026: the state of play" - Edward Jansen
18:30 – 19:10 | "Red teaming your AI agent before it ships" - Bauke Brenninkmeijer
19:10 onwards | Walking dinner and networking
20:00 | Closing
Location: ADC, Amsterdam office, De Ruijterkade 7, 1013 AA Amsterdam
Get in touch
Questions about the event? Contact Jamie Couchman at james.c@adc-consulting.com