Insights Hub
Stay ahead with perspectives on real-time analytics, streaming architectures, and industry trends.
Latest Insights
Articles
Read-time current-state: how ISK deduplicates CDC on Iceberg
Diving into the mechanism behind ISK: how it transforms a Kafka CDC stream into a deduplicated, current-state Iceberg table that any Iceberg-compatible engine can query.
July 17, 2026 · By Roman Kolesnev
Read More
Articles
The freshness gap: uniting real-time and batch analytics
Near-real-time CDC analytics costs are driven far more by continuously reconciling and maintaining current-state tables than by data volume itself, and Streambased argues that decoupling freshness from commit cadence can reduce those costs by an order of magnitude while simplifying architecture.
July 8, 2026 · By Roman Kolesnev
Read More
Articles
The cost structure of near-real-time CDC analytics
Near-real-time CDC analytics costs are driven far more by continuously reconciling and maintaining current-state tables than by data volume itself, and Streambased argues that decoupling freshness from commit cadence can reduce those costs by an order of magnitude while simplifying architecture.
July 1, 2026 · By Roman Kolesnev
Read MoreAll Insights
Three problems, one solution
Streambased unifies Kafka and Iceberg into a single, queryable view, live and historical data together, without moving or copying anything. That one capability can answer a wide range of different data challenges.
August 31, 2026 · By Leo Delmouly
How Streambased powers AI in property and facility management
Why does AI so often fall short in property and facility management? Our latest blog explores the hidden data context problem.
August 4, 2026 · By Dean Ward
The right care at the right time: how Streambased powers AI in healthcare scheduling
In healthcare, the cost of acting on incomplete context is measured not in sales but in patient outcomes. This is the fifth post in our six-part series on Streambased and AI.
July 29, 2026 · By Dean Ward
From alerts to answers: rethinking AI anomaly detection
Detecting an anomaly is easy. Determining whether it matters - and what caused it - is where AI succeeds or fails, depending on the quality of the data it can access.
July 22, 2026 · By Dean Ward
The data giants just doubled down on yesterday’s architecture
The article argues that the future of data systems is accessing data directly when needed instead of constantly copying it between systems, making data simpler, cheaper, and more up to date.
June 23, 2026 · By Tom Scott
AI personalisation
Someone’s browsing running shoes on your site – but are they a casual jogger or a serious runner? The answer completely changes what you should recommend next, but most personalisation models can’t tell the difference. Streambased can.
June 22, 2026 · By Tom Scott
Seeing the full customer with AI requires context
AI without context can't deliver good support to customers - it can't see the right signals without both historical context to back up real-time data
June 11, 2026 · By Tom Scott
What your AI agents don’t know is costing you
Most AI agents don’t fail because of the model but because the model is working with an incomplete picture. In this six-part series, we look at how Streambased gives AI the context it needs to reflect the business as it actually is, and all the use cases that unlocks.
May 25, 2026 · By Tom Scott
See you at Current London?
If you work with Kafka, Iceberg or any part of the modern data stack, Current London on 19-20 May at ExCeL is the event to be this spring. Two days, 2,000-plus attendees, all the thought leaders in data streaming and the modern stack... This is the community that’s shaping where the ecosystem goes next. We’re there as silver sponsors, our CEO Tom Scott is giving two talks, and we’d love to see you there.
April 30, 2026 · By Dean Ward
ETL moves data. It can’t make better decisions
Businesses have spent years building data pipelines to close the gap between their operational systems and their analytical tools. Yet still decisions get made based on incomplete, lagged data. That’s because no matter how fast your pipes, moving data is not the same as unifying it – and the disconnect is costing businesses dear...
April 17, 2026 · By Dean Ward
Choose Between Speed vs Context in Modern Data Architecture
Real-time systems give you speed but no history.
March 11, 2026 · By Dean Ward
Streambased at Current New Orleans 2025
Join us for: Turning the database inside out again: What if everything was Iceberg?
October 29, 2025
The 9 Ways to Move Data Kafka to Iceberg
An in Depth comparison from the Streambased ZeroCopy blog.
October 26, 2025
Streambased at Data Decoded Manchester
We’re looking forward to two days of great conversations around real-time data, Apache Kafka, and Apache Iceberg, and to meeting the brilliant Manchester tech community in person.
October 21, 2025
Vibe Coding for Data at Data Streaming Summit San Francisco
Join us in San Francisco at the end of the month for Data Streaming Summit where we'll be using AI and real-time data to solve a crime live!
September 30, 2025