LDPaC 29th Sept 2026: Leed Digital Festival Week

This meetup is an exciting opportunity for us here at LDPaC - we are hosting an event in conjunction with Leeds Digital Festival! We have two really exciting talks from some well versed speakers, focussing on Agentic AI and AI Red Teaming and security vulnerabilities.

Sign up here to attend this exciting event!

*In Person Only MeetUp, Free To Attend*

Thank you to Accenture for sponsoring and hosting this event.

Join us for a In-Person User Group Meeting (LDPaC), where you can network, learn, ask a question, and meet other likeminded folks. These events are a really great opportunity to socialise in an informal learning setting.

Remember to tell your friends and the people you work with; make sure you register as soon as you can. We will need to provide a list of names to Accenture before the event, so to ensure there are no issues with access on the day please make sure you have registered.

18:00 - 18:30 Network🤝/ Drinks / Pizza 🍕

18:30 - 18:40 Intro🎙️

18:40 - 19:30 Talk 1: Shubhangi Goyal - Context driven AI agents

In this session, I will explore how agents are built and use dynamic context to execute complex tasks and deliver real value. This also includes using context engineering to move beyond single prompt based interactions. The key takeaways would be :

- Context engineering, manage memory tools, and tasks for adaptive behaviour
- Agent architecture, LLMs and feedback loops
- Prompt engineering into the larger system
- Real world use case and common pitfalls when deploying AI agents

19:30 - 19:35 5-min Break 🥤

19:35 - 20:25 Talk 2: James Bentley - Red Team Redemption: Finding AI Weaknesses Before Attackers Do

Generative AI is rapidly becoming part of enterprise applications, but how do you know if your AI behaves safely when faced with malicious, unexpected, or simply creative users?

In this session, we'll explore the world of AI red teaming: the practice of thinking like an attacker to uncover vulnerabilities before they reach production. We'll look at common attack techniques and adversarial behaviours which can cause AI systems to act in ways their creators never intended.

Using real world engagements as inspiration, you'll see how open source security tools such as Garak, PyRIT, and more can be combined into a repeatable testing framework, and how these findings map to industry standards to make sure you know where your gaps are, and what that even means!

Of course we'll also discuss the practical mitigations you can implement, and the growing security capabilities available within Databricks to safeguard your deployments.

So whether you're deploying your first chatbot, or operating AI at enterprise scale, this session will help you understand how attackers think, how AI systems fail, and how to ensure your first red team engagement isn't conducted by your users.

20:25 : 20:30 Wrap Up & Thank you's

Come and join the Leeds Data Community and start learning and networking! All are welcome!