Program JavaZone

Workshop Tirsdag 1. september

From Streams to SQL to AI. Build an Open Source HTAP Platform in 4 Hours

Workshop A

Engelsk 4 t

Saleem Pothiwala

28+ years in data. Built Kermit Tech to help Nordic and Middle East enterprises turn data into competitive advantage.

Specialized in:

• RAG & AI implementations for finance, oil & gas, retail
• IBM watsonx.data integration and watsonx.data intelligence platforms at scale (250+ users)
• Cloud migration (on-prem to Azure/AWS/GCP)

Led teams at DNB, Shell, Virgin Media. Now running a consultancy with offices in Norway, UK, and UAE.

IBM Champion 2025, 2026. Data Engineering Professional of the Year 2024 (Nordics).

Modern AI and analytics workloads punish architectures built on copying data between systems. In this workshop, we build a vendor-neutral, open-source HTAP stack that combines transactional, analytical, and AI-oriented workloads without the usual ETL tax. The architecture is based on a practical demo stack using Apache Cassandra for operational data, Apache Kafka for ingestion, Apache Spark for large-scale processing, Presto for interactive SQL, and open table and search technologies such as Parquet, Iceberg, and OpenSearch.

Participants will bring up the stack locally, stream events into it, query live operational data, run analytical queries, and export data for downstream use. Along the way, we will discuss why real-time context, cost control, and reducing duplication matter for modern systems, especially when building AI-ready platforms and agent-facing applications.

This is not a product pitch. It is a hands-on architecture workshop for developers and architects who want to understand how open-source building blocks fit together to support low-latency apps, analytics, and AI use cases in one platform.

Java developers working on data-intensive or event-driven systems. Data engineers and platform engineers. Solution architects evaluating modern data platform patterns. Teams exploring AI-ready data architectures without vendor lock-in.

Participants should prepare the following before the workshop: 1. A laptop with Docker Compose or Podman Compose installed. 2. At least 12 GB RAM available for the local environment for bringing up the full stack locally 3. Git installed 4. A terminal and an editor or IDE of choice 5. Optional but useful: Java and basic command-line comfort Before the session, participants should clone the workshop repository and verify they can start containers locally.

  1. Pleesah, lær meg mer om Kubernetes! Workshop B
  2. AI Can't Debug This: A Virtual Thread Migration Workshop Workshop C
  3. Bygg riktig før du bygger ferdig: Tjenestedesign for utviklere Workshop D
  4. Event Sourcing: From Chaos to Control Workshop B
  5. Build AI Agents That Actually Do Things — Hands-on with MCP Workshop C
  6. From data to OGC API (Features) service using Geoserver Workshop D