Tom Liptrot - Agentic Memory

Join us for another Practically AI Manchester as we explore one of the most important challenges facing modern AI systems: memory.

As AI agents become more capable, one question keeps surfacing: what should an agent remember?

What should it forget? How do memories evolve over time? And how do we stop agents accumulating irrelevant or misleading context that impacts performance?

Our featured session will be delivered by Tom Liptrot, who will explore the practical realities of agentic memory. Drawing on experience building and working with AI systems, Tom will examine how memories are formed, stored, retrieved, and updated within agent-based architectures. Along the way, he'll compare these approaches with human memory, highlighting both useful parallels and the ways AI systems remain fundamentally imperfect.

Alongside Tom's session, we'll also be covering some of the most talked-about topics in AI right now, including:

• AI Spotlight
• How AI is Affecting the UK Job Market and How to Position Yourself
• AI Hub Update: Eliza 3.0

Whether you're building AI systems today, exploring agentic workflows, leading engineering teams, or simply trying to stay ahead of the rapid pace of change, there should be something for you.

Agenda

17:30 – Doors open, pizza & networking
18:15 – Welcome & introductions
18:20 – AI Spotlight
18:35 – How AI is Affecting the UK Job Market and How to Position Yourself
18:50 – AI Hub Update: Eliza 3.0
19:10 – Talk: Agentic Memory – Tom Liptrot
19:50 – Q&A
20:00 – Open social, networking & refreshments
20:30 – Close

Venue

BNY, 1 Piccadilly Gardens, Manchester M1 1RN

Entry

Upon arrival at BNY, please check in at reception and you'll be directed to the meetup.

Please note, attendees must show a form of ID on entry.