Current London 2026 - Register Now!

The biggest data streaming event of the year is back! Register now for Current London 2026 and witness 100+ eminent speakers deliver 80+ sessions on the cutting edge of Data and AI technologies!

The grid is set for London.

In the world of data, "lights out" isn't just a start time, it’s the standard for latency. We’re bringing Current26 to the ExCeL London on May 19–20, and the stakes have never been higher.

Just like an F1 race, successful data streaming requires more than just a fast engine (Kafka). It needs the right strategy (Governance), real-time adjustments (Flink), and a world-class pit crew (you).

Don't get stuck in the pits while the industry moves to real-time.

Secure your pass: https://bit.ly/4tQbXdD  

Register Now - Current London 2026

New with Confluent Platform 8.2: Queues for Kafka, Flink SQL (GA), updates to CPC Gateway, and more..

We’re excited to announce Confluent Platform 8.2!

The latest Confluent Platform release, built on Apache Kafka® 4.2, extends and simplifies what you can use Apache Kafka® and Apache Flink® for, whether that’s handling task queues natively, processing streams with SQL, or managing cluster migration, upgrades or disaster recovery without the usual operational pain.

What’s new in Confluent Platform 8.2

The latest release of Confluent Platform enables organizations to:

• Run streaming and task-queue workloads side by side with Queues for Kafka, with native queue semantics and elastic consumer scaling built in.
• Simplify stream processing with Flink SQL (GA): Filter, join, aggregate, and transform data streams in Confluent Platform for Apache Flink® using DDLs, changelogs, and shared compute pools providing a declarative way to manage Kafka topics directly through the Confluent CLI or the Control Center UI.
• Reduce operational complexity with Flink ease-of-use enhancements, including multi Kubernetes cluster support, a new savepoint management UI, and native support for RHEL 10 and OpenShift environments.

Available with Confluent Private Cloud

Confluent Private Cloud (CPC) Gateway allows you to decouple applications from your infrastructure. The latest release of CPC Gateway 1.2 includes SCRAM auth swapping and client fencing, making it easier to manage migrations without touching application code.

Read the blog to learn more

Data Streaming Resources

• Confluent Cloud for Apache Flink now supports configurable late data handling rules. Use the new late-handling.mode table property to drop or pass-through late-arriving events, and/or use a $late system table to send these late events for reconciliation.

What's New?

In streaming systems, "late data" refers to events that arrive after the system's watermark has advanced past their timestamp. Until this change, late events were mostly silently dropped.

The Custom Late Data Handling feature gives users full control over those records:

• Simple on/off control: Set late-handling.mode = 'filter' to separate late records from the main pipeline (default is pass-through, preserving existing behavior)
• System Tables: Late records are automatically captured in a <table_name>$late system table. This table has the same schema, full access, no additional setup required
• Easy configuration: Apply at table creation with CREATE TABLE, or update any existing table at any time with ALTER TABLE

Why is this important?

• No more silent data loss: Late records are preserved and available for reprocessing or audit, regardless of the main pipeline's watermark strategy
• Correct aggregations: Window operations and time-based joins produce accurate results without late stragglers skewing output
• Flexible reconciliation: Teams can UNION the main stream with $late to account for every event.

Getting Started

Query the late records at any time:

SELECT * FROM `my_table$late`;

Enable filtering on an existing table to drop them at the source

ALTER TABLE my_table SET ('late-handling.mode' = 'filter');

Check out the docs for full configuration options, reprocessing patterns, and monitoring guidance.

• Read this fantastic blog from the Notion engineering team on how Notion has implemented Data Residency for their customers using a modular and agile multi-region infrastructure with Apache Kafka being the foundation for event logging.

Links From Around the Web:

• The DuckDB team has released DuckLake v1.0, a production-ready lakehouse format specification built on SQL. A key difference between other formats and DuckLake is that DuckLake stores all metadata in a database, which is commonly referred to as the catalog. This database can be any system that speaks SQL, supports primary keys and is able to persist data in tables. The DuckLake specification defines the metadata tables needed for a specific version along with the supported data types and a reference on how to retrieve table metadata to perform operations on the lakehouse.
• Corey Quinn takes developers on a wild ride while testing READ and WRITE processes of AWS S3 Files, which was released a week back in April, 2026. Read his experience, here.

In-Person Meetups

• May 6 - Phoenix w/Elastic
• May 7 - London @ Fresha
• May 7 - Belgium w/Cymo
• May 12 - Sydney
• May 13 -Chicago w/Elastic
• May 16 - Coimbatore
• May 16 - Bangalore
• May 21 - Madrid @ MasOrange
• May 21 - Singapore @ IMDA
• May 25 - Toronto @ Improving
• May 26 - Erlangen/Nurnberg w/Siemens Healthineers
• May 27 - Minneapolis w/Elastic
• May 27 - Sydney w/ Elastic
• May 28 - Lisbon

By the way…

We hope you enjoyed our curated assortment of resources! If you’d like to provide feedback, suggest ideas for content you’d like to see, or you want to submit your own resource for consideration, send us an email at devx_newsletter@confluent.io!

If you’d like to view previous editions of the newsletter, visit our archive.

If you’re viewing this newsletter online, know that we appreciate your readership and that you can get this newsletter delivered directly to your inbox by filling out the signup form on the left-hand side.

P.S. If you want to learn more about Kafka, Flink, or Confluent Cloud, visit our developer site at Confluent Developer.