[ANNOUNCE] Spark on Kubernetes Community Calls

[Andrey Velichkevich]

Hi Kubeflow Community,

Happy New Year 🎉

We’re excited to announce the launch of our bi-weekly Friday community calls focused on running Apache Spark on Kubernetes.

📅 The next meeting will be held on January 9th 2026, 9:00am PST.

• Meeting Notes & Agenda
• Register via the Kubeflow LFX calendar

Deploying large-scale data processing workloads on Kubernetes continues to present real challenges. These calls are designed to create a collaborative space where the community can share real-world challenges, demos, lessons learned, and provide direct feedback to Kubeflow maintainers.

During these meetings, we plan to discuss a wide range of topics and projects, including the Spark Operator, the MCP Server for Spark History Server, Apache DataFusion and Apache Iceberg in Kubeflow Trainer, and other efforts to enhance data management on Kubernetes. We will also explore how to accelerate Spark using technologies such as DataFusion Comet and Apache Gluten with Velox.

High-quality data is the foundation of modern AI systems and LLMs. We believe data workloads are just as critical to running AI on Kubernetes as training, inference, and agentic workloads, and deserve the same level of focus and innovation.

For more details watch the recording from our last Spark Operator call.
You can also find us and ask questions in the CNCF Slack channel: #kubeflow-spark-on-kubernetes
hashtag

We invite data engineers, platform engineers, and AI practitioners to join us and help shape the future of Data & AI on Kubernetes!

Regards,

Andrey