AI Summit at a glance
The first Zartis AI Summit held place on November 17th, in Madrid, Spain.
The goal of the event was to explore the various applications of AI within technology companies by bringing together technology leaders and experts in the field.
Interesting Keynotes
Keynote - Overview of AI Regulatory Landscape
Keynote speaker: Dr Florian Ostmann, Head of AI Governance & Regulatory Innovation, The Alan Turing Institute
Keynote - Delivering Business value with AI
Keynote speaker: Federico Castanedo, Academic Director and Professor at IE University
Interactive Panel Discussions
Panel - AI & Regulatory Considerations
Karl Aherne, COO at Fexco
David Boundy, COO at Innowatts
Darren Hayes, VP of Engineering at valid8Me
Phil Thomas, Tech Head of Digital at Zartis
Panel - AI-Driven Operational Efficiency
Jim Carr, Director of Engineering at CarTrawler
Ciaran Bradley, CTO at EclecticIQ
Angel Benito, CTO at Zartis
Tech Showcases
Zartis AI Committee - Lessons Learned
Insights shared by the Zartis AI Committee on AI tools selection, GenAI development, and implementation.
Insights on:
- Setting up an AI strategy
- AI-powered programming tools
- AI-powered QA and testing tools
- AI-powered DevOps tools
AI insights from the summit:
Large context windows solved capacity but birthed 'context rot.' Discover why frontier models degrade over long workflows, and how the 'lost in the middle' effect corrupts agent reasoning.
Learn why LLM inference is limited by memory bandwidth, not compute. Explore a 6-layer optimization stack to achieve 4–20x gains in throughput and cost.
Stop treating AI as a series of pilots. Learn how to build an enterprise AI operating capability that integrates people, platforms, and governance for scale.
Learn how AI for power grid systems must shift from probabilistic models to reliability engineering to support critical infrastructures.
Unlock a practical AI readiness assessment for CTOs in regulated sectors. Bridge the gap between technical pilots and institutional trust.
Stop settling for better forecasts. Discover how AI supply chain optimization, digital twins, and autonomous decision-making are replacing static planning to build resilient, real-time logistics networks that thrive on volatility.
Discover why the standard Kubernetes playbook fails for LLM workloads and get a production-ready strategy for scaling vLLM, KServe, and Triton.
Learn how to bridge the "AI Governance Gap," solve the problem of "Orphaned Agents," and implement the 4 structural capabilities needed for responsible AI autonomy.
Learn the five architectural patterns - from context avalanches to damped retry loops - for calculating the only metric that matters: reliability-adjusted cost per task.
Individual agent performs well, but the pipeline collapses anyway. Here's the mathematics of why, and what a real error-reduction architecture looks like.
Let's discuss how you can unlock the full potential of AI