Real-Time Data & AI: Scale, Speed, and Insights

Please note: Registration for this event can be completed [here].

Join us for an evening focused on tackling the real-world challenges of modern data systems. This isn’t just theory - it’s a deep dive into database architecture, real-time streaming, and AI-ready infrastructure that delivers speed, scale, and actionable insights.

We’ll bring together experts from Aerospike, Seznam cz, and Principal Engineering to share practical lessons from production systems, migration journeys, and real-time data pipelines.

Learn how to evaluate databases for AI workloads, scale mission-critical systems, and turn messy streams into structured, query-ready data using Kafka, Flink, and Iceberg.

What you’ll learn:

• Scaling data infrastructure while maintaining robustness, low latency, and high reliability.
• Evaluating databases for AI-driven workloads using a vendor-neutral framework focused on real-time performance.
• Turning real-time data streams into actionable insights with Kafka, Flink, and Iceberg.

Whether you’re a data engineer, architect, or developer, you’ll walk away with actionable ideas for building systems that are fast, scalable, resilient, and ready for AI applications.

Agenda
17:30 - 18:00 - Drinks and Networking

18:00 - 18:15 - Databases at the Crossroads of Scale, Real-Time, and AI
Zohar Elkayam, Solutions Architect, Aerospike

This session is designed for intermediate to advanced practitioners and dives deep into the technical nuances of database architecture and evaluation in today’s AI-driven era. We’ll examine how teams often end up with systems that don’t align with their actual workload needs, exploring critical dimensions like consistency models (eventual vs. strong), scalability (vertical vs. horizontal, single vs. multi-region), access patterns (key-value, document, graph), and durability trade-offs (synchronous vs. asynchronous replication). As AI workloads bring unprecedented demands for low latency, massive scale, and adaptive infrastructure, the importance of selecting the right database has never been greater. Rather than advocating for a single technology, the talk presents a practical, vendor-neutral framework to evaluate and compare database types. Through real-world examples, we’ll highlight common anti-patterns, share lessons from successful migrations, and show how the right database choice can dramatically improve performance, resilience, and team efficiency.

18:15 - 18:40 - Seznam cz and Its Journey with Aerospike: Early Lessons from Migration
Michal Fizek, Development Team Leader, Seznam

We are migrating from the community edition to the enterprise edition of Aerospike. This talk offers a practical overview of the early challenges we’ve faced - both operational and architectural - and what we’ve learned so far.
We’ve been running Aerospike in production for 10 years, primarily to store session data and registered-user profiles. Because this system directly powers the logged-in user experience and underpins business-critical adtech, robustness, security, and low latency are absolutely essential. The infrastructure also processes GDPR-sensitive data, making dependability and safety non-negotiable. As we transition to the enterprise edition, we are beginning to see how its features help us meet these requirements.

18:40 - 19:05 - Mission-Critical Policy Control (PCRF/PCF): Achieving Five Nines of Availability
Jan Čaboun, Senior Project Manager, Principal Engineering

In mission-critical environments, even a few seconds of downtime can have a massive impact. In this talk, we’ll share how we built and operate a Policy Control (PCRF/PCF) platform that achieves five nines (99.999%) availability - combining real-time performance, reliability, and peace of mind in production.

19:05-19:30 - Impact First – How to Develop at Scale and Speak in Metrics
Odelya Holiday, Staff Software Engineer, Distributed Systems Engineer

In this session, Odelya will share a practical framework for thinking about engineering tasks through the lens of measurable business impact. She’ll walk through how she approached a real-world challenge, reducing user churn, by implementing a Kafka-based retry model and building a data funnel to monitor results and track impact over time.

Beyond the technical solution, the session will explore the importance of communication and visibility, how to translate technical work into metrics that resonate across teams and clearly demonstrate organizational value. This talk is ideal for engineers looking to connect their technical contributions with broader business outcomes.

19:30 - 20:30 - Networking