IN PERSON: Apache Kafka® x WarpStream x MCP

Join us for an Apache Kafka® x WarpStream meetup on Wednesday, October 15th from 6:00pm hosted by Fiberplane!

📍Venue:
Raadhuisstraat 50, 1016 DG Amsterdam, Netherlands

🗓 Agenda:

• 6:00pm: Doors open
• 6:00pm – 6:30pm: Drinks and networking
• 6:30pm - 6:55pm: Jacco Flenter, Frontend Engineer, Fiberplane
• 7:00 pm - 7:25 pm: Dani Torramilans, Staff Software Engineer, Confluent
• 7:30pm - 7:55pm: Debabrata Panigrahi, Founding Engineer, Parseable
• 8:00pm - 9:00pm: Food/Additional Q&A and Networking

💡 Speaker One:
Jacco Flenter, Frontend Engineer, Fiberplane

Title of Talk:
What about MCP?

Abstract:
The Model Context Protocol (MCP) is a new standard for connecting AI assistants to external systems. In this talk I’ll share a developer’s perspective on where MCP could be useful, the kinds of challenges that come up when building servers, and how we’re approaching them at Fiberplane.

💡 Speaker Two:
Dani Torramilans, Staff Software Engineer, Confluent

Title of Talk:
Building on the Shoulders of Giants: How Cloud Primitives and Hyperscalers Simplify Distributed Storage

Abstract:
Distributed databases have always been hard to build. Today, cloud primitives such as durable object storage and single-digit millisecond key-value stores are reshaping that landscape. By abstracting away many of the hardest problems in distributed systems, they enable simpler, more maintainable implementations. This shift is driving the rise of Kafka-on-S3 systems and data lake query engines.

In this session, we’ll look at WarpStream, a petabyte-scale Kafka replacement built entirely on cloud primitive. Together, we’ll explore:

• Core concepts of building databases with a clean separation of storage and compute, and how cloud primitives make advanced features easier to deliver and operate.
• WarpStream’s design: stateless, diskless agents for the data plane and stateless, multi-tenant instances for the control plane, with zero dependency on local disks.
• Trade-offs between traditional disk-based systems like Kafka and Elasticsearch compared to cloud-native databases, focusing on performance, latency, and cost.
• How cloud-native design accelerates features such as multi-region deployments and integrated data warehousing.

💡 Speaker Three:
Debabrata Panigrahi, Founding Engineer, Parseable

Title of Talk:
Making sense of Kafka metrics with Agentic design

Abstract:
In this talk we will look at how to track and export Kafka metrics from a Kafka production cluster to an observability system like Parseable. We'll then deep dive into the metrics data, its implications and more. Finally we'll look at a LLM based agentic style workflow to see how to predict metrics data points, set up relevant alerts and create actionable insights from this metrics data.

Bio:
Bio: Debabrata Panigrahi is a founding engineer at Parseable, building a unified observability platform. Previously led DevRel at Harness for the Internal Developer Portal; OSS contributor to Kubernetes, Chaos Mesh, and Backstage.

***
If you are interested in providing a talk/hosting a future meetup, please email community@confluent.io