What Happens When AI Leaves the Lab

Frontier AI models are evolving quickly, but the real challenge is no longer just building larger systems, it is turning advances in LLM research into reliable, efficient, and valuable products in the real world.

We invite you to join a panel discussion featuring experts working across AI research and engineering, who will share their perspectives on what actually matters in the next wave of LLM development. The conversation will explore the trade-offs between scale, efficiency, reasoning, and product impact, as well as what it takes to move from experimental models to systems that can be deployed at scale. The discussion will be moderated by Rogers Leo, Director of Engineering and Product (Agentic AI) at Mews.

This panel will go beyond the hype and focus on practical lessons from building with LLMs today. We will look at where current approaches break down, what “good enough” really means in production, and why efficiency and system design may matter as much as model capability.

Panelists:
Karol Lasocki, Senior Machine Learning Engineer at BottleCap AI
Luděk Matyska, AI Lead at ROUVY
Soheyla Mirshahi, AI Governance & Product Leader at T-Mobile

Agenda:
18:00 Doors open
18:30 Panel discussion + Q&A
19:30 Networking

Photographs and video recordings may be taken during this event for documentation and promotional purposes, including use on websites, social media, and other communication channels. If you do not wish to be photographed or recorded, please inform a member of the event staff upon arrival or at any point during the event, and we will ensure your preference is respected.