From the Desk of the Confluent Developer Newsletter Team

Dear Data Streaming Community,

As we look back on the year gone by, it is impossible not to be amazed at the speed at which AI (especially Generative AI) has infiltrated every sphere of our daily lives! 2025 has been a watershed year where large language models (LLMs) reigned supreme and the data community as a whole experimented with novel ways to integrate this powerful technology.

Data streaming has also seen major progress with engineering teams across the globe adopting streaming technologies, such as Apache Kafka® and Apache Flink® to integrate and augment their data pipelines with AI.

‌In this edition of the newsletter, let's look back at major releases and events that took place throughout 2025!

‌Also, we sign off with 11 newsletters published in 2025! You can revisit an old edition at our usual Confluent Developer Newsletter section! We can't wait to publish another year full of newsletters chronicling the journey and evolution of data streaming technologies in 2026!

‌Stay safe, and wishing you a very happy and prosperous 2026 ahead!

‌P.S.: You can now subscribe and be notified of updates to the Confluent Cloud release notes using an RSS feed reader! Subscribe to Confluent Cloud product feature updates on RSS feeds!

Apache Kafka and Apache Flink Highlights from 2025:

• The biggest highlight of 2025 for the Kafka community has been the introduction of Apache Kafka® 4.0! Apache Kafka 4.0 was a significant milestone, marking the first major release to operate entirely without Apache ZooKeeper™. By running in KRaft mode by default, Kafka simplifies deployment and management, eliminating the complexity of maintaining a separate ZooKeeper ensemble. This change significantly reduces operational overhead, enhances scalability, and streamlines administrative tasks. This Confluent blog summarized most of the KIPs across the Producer, Consumer, Admin Client, Kafka Streams and Kafka Connect.
• The Kafka community released Apache Kafka® 4.1.0 with a preview of Queues for Kafka. This Confluent blog summarised the release features.
• Apache Flink® 2.0 was also released in 2025 and introduced Disaggregated State Storage and Management, Materialized Tables among other exciting features. The release note from the Flink website summarized the new features.
• The community also witnessed consistent releases of librdkafka, which powers Confluent’s official Python, Go, .NET and JavaScript clients as well as many more community clients! The most recent version delivers on stability and parity, incorporating critical features and resolving 11 GitHub issues. This release incorporates direct feedback from the community: 
  • Security & Reliability: Fixed a critical double-free issue for increased stability and security.
  • Resilience: Further refined the KIP-1102 time-based re-bootstrap condition to ensure smarter, more effective metadata recovery.
  • Consumer Robustness: Fixed an issue where topic partition errors were not cleared after a successful commit, guaranteeing accurate error reporting.
  • API Reliability: Fixed an issue to ensure the specified timeout is correctly enforced.

Event Highlights from 2025

• Current, the biggest data streaming event had three stops in 2025 - Current Bengaluru in March, Current London in May and Current New Orleans in October 2025.
• Current remains the biggest event where the data streaming and data engineering communities meet and exchange ideas, with thousands of data enthusiasts visiting the event, listening to major feature announcements and attending breakout sessions.

Data Streaming Platform Updates

• Confluent Intelligence launched at Current NOLA – providing a fully managed service on Confluent Cloud to help build intelligent real-time AI systems using Apache Kafka® and Apache Flink®:

• Get started with out-of-the-box Anomaly Detection, Fraud Detection, Forecasting, and more with built-in ML functions – no ML expertise required.
• Build your first Streaming Agent in minutes, delivering event-driven, real-time agents that can monitor and instantly act on live operational events.
• Sign up for Early Access to the Real-Time Context Engine to empower any AI agent and application anywhere to make informed decisions based on the freshest context.

Current Bengaluru and London 2026: The road ahead

Submit papers for Bengaluru ➡️ Sessionboard
Submit papers for London ➡️ Sessionboard

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, email us 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 sign-up form below.

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