One platform for real-time, batch and ad-hoc analytics
We design and build open lakehouse platforms that combine streaming, batch processing and interactive SQL on the same data. They run on Kubernetes in your data centre or in the cloud, and use open formats, so you are not locked in to a single vendor.
Our teams have run Kafka, Spark, Flink and Kudu in production for capital markets and aviation, and we modernise legacy data warehouses and Hadoop clusters to the lakehouse.
What we deliver
Lakehouse architecture
Apache Iceberg tables on object storage, with Spark for batch, Trino for ad-hoc SQL and Flink for streaming, all on one copy of the data.
Real-time streaming
Kafka and Apache Flink pipelines for live dashboards, alerts, fraud detection and event-driven applications, with exactly-once processing.
Warehouse modernisation
Move from legacy MPP warehouses, Hadoop or Cloudera clusters to an open lakehouse, step by step and without stopping existing reports.
Semantic layer and BI
Governed business definitions with dbt and Dremio, served to Power BI, Qlik Sense and MicroStrategy.
Data integration
Batch and change-data-capture ingestion from Oracle, SQL Server, PostgreSQL, APIs and files, orchestrated with Airflow or NiFi.
Operations and support
Monitoring, capacity planning, upgrades and performance tuning for the platform after go-live.
The building blocks we use
How a platform project runs
Assessment
We review your sources, data volumes, latency needs and current reports, and agree the target architecture.
Foundation
We set up storage, compute, security and the first pipelines on Kubernetes, in your environment.
First use cases
We move or build the first high-value reports and real-time flows, and run them in parallel with the old system.
Scale and hand over
We migrate the remaining workloads, train your team and support the platform in production.
Enterprise support when you need it
Ververica
As a Ververica partner, we offer the enterprise Apache Flink platform from the original creators of Flink, with vendor support for critical streaming workloads.
Cloudera
For organisations running Cloudera, we support the existing platform and plan a gradual path to an open lakehouse.
Planning a new data platform or a migration?
Share your current architecture and goals. We will propose a target design and a phased plan.