We are thrilled to announce a milestone that has been one of our top customer requests: DeltaStream is now available for Early Access (EA) on Microsoft Azure.

Data teams operating in the Azure ecosystem can build, manage, and scale sophisticated, stateful streaming data pipelines from end to end using only SQL. This release is a major step in our mission to make powerful stream processing simple, accessible, and truly serverless—no matter where your data lives.

The Real-Time Challenge in the Azure Ecosystem

The Microsoft Azure stack is a powerhouse for data. Teams rely on Azure Event Hubs for high-throughput data ingestion and Azure Data Lake Storage (ADLS) Gen2 for building massive, cost-effective data lakes. Combined with services like Azure Synapse Analytics and Microsoft Fabric, it’s a world-class analytics platform.

But a critical gap exists: the real-time processing layer that connects ingestion to analytics.

Until now, data teams faced a difficult choice:

1. Manage Complex Clusters: Use powerful but operationally heavy platforms like Apache Spark or Flink. This requires a dedicated team of specialized engineers, high costs for idle clusters, and a steep learning curve.
2. Default to Batch: Give up on real-time. Teams are forced to run batch ETL jobs every 15 minutes, 30 minutes, or even hourly, delivering stale data to the business.

We believe Azure users deserve a better option. You shouldn’t have to be a distributed systems expert to build a real-time medallion architecture.

DeltaStream: The Serverless Streaming Layer for Azure

DeltaStream is the serverless, SQL-native stream processing platform that bridges this gap. We provide the power of a complex streaming engine with the simplicity of SQL, all delivered through flexible deployment models that fit your needs.

• 100% SQL-Native: If you and your team know SQL, you can now build and manage robust streaming pipelines. Create streams, tables, and materialized views using the language you already know. No Java, Scala, or Flink expertise is required.
• Powerful Stateful Processing: Go far beyond simple ETL. Perform time-windowed aggregations (e.g., “count clicks by user in the last 5 minutes”), join your real-time Event Hub stream against a reference table, and materialize those results for downstream consumption.

Flexible Deployment: SaaS or Your Own VNet (BYOC)

We understand that every enterprise has different security and governance requirements. That’s why DeltaStream on Azure is available in two models:

• SaaS (Software-as-a-Service): The simplest way to get started. This is our fully managed, multi-tenant environment. You get all the power of DeltaStream without managing any infrastructure—just connect to your Azure sources and sinks, and we handle the rest.
• BYOC (Bring Your Own Cloud): For organizations with strict data residency and security needs. DeltaStream deploys its serverless data plane directly inside your Azure subscription (in your VNet). This ensures your data never leaves your network boundary, giving you the full security and governance of your own environment, but with the simplicity of a managed serverless platform.

This flexibility empowers all Azure customers—from fast-moving startups to security-conscious enterprises—to build the real-time use cases they’ve always wanted with a fraction of the cost and complexity.

Join Our Early Access Program

We are now opening our Early Access program for a select group of Azure customers.

EA participants will be the first to build production-grade streaming pipelines on Azure with DeltaStream, and they will work directly with our founding engineering and product teams to help shape the future of our Azure integration.

Ready to stop managing clusters and start building with SQL?

Request early access: https://www.deltastream.io/azure-early-access-program/

New features include a Real-time Context Engine for GenAI Agents and MCP Server—uniting AI-assisted pipelines, unified orchestration, cost optimization, and observability in a single serverless architecture

MENLO PARK, Calif. — October 29, 2025 — DeltaStream, Inc., the serverless stream processing platform, today announced the broader availability of DeltaStream Fusion, its unified streaming data platform that brings streaming, real-time, and batch analytics together in a single, serverless experience. Originally launched in early access, DeltaStream Fusion is now production-ready and validated by customers—and is also available on Microsoft Azure, giving organizations even greater flexibility to deploy where their data lives. With enhanced AI-driven features and new product capabilities, including the ability to power GenAI applications and agents with real-time context, DeltaStream empowers enterprises to unlock real-time intelligence from their data, unify analytics pipelines, accelerate time to insight, and reduce operational costs.

Today’s organizations are under pressure to harness data for real-time decisions, AI, BI and business operations. But their architecture is working against them. Many rely on separate systems for streaming, real-time and batch analytics, which creates fragile pipelines, integration overhead and higher operational costs as data volumes grow and compute storage explodes. At the same time, batch-heavy pipelines lead to stale data, while engineers prioritize provisioning, tuning and scaling infrastructure rather than enabling faster insights. And, in an effort to address these issues with streaming-native tools, some organizations get stuck trying to demystify open-source stream processing technologies. 

While open-source point products are hard to develop with — and harder to scale and operate — DeltaStream Fusion, a unified streaming data platform, brings together streaming, real-time and batch analytics in a single, serverless experience. It is architected from the ground up to deliver trustworthy and contextual insights in real time, and it removes the engineering burden of managing infrastructure. And, it does not require specialized skills; rather, DeltaStream Fusion lets any team write once in SQL and automatically selects the optimal engine, enabling both performance and agility. Key innovations include:

• AI-assisted pipeline creation with simple SQL, so teams can go from raw data to action in minutes
• Unified engine orchestration across Apache Flink, Spark and ClickHouse; no need to manage multiple tools
• Serverless architecture that eliminates infrastructure overhead while scaling instantly
• Cost optimization that helps reduce exploding Snowflake and Databricks bills by shifting analytics left
• Built-in observability and governance for enterprise-grade reliability and compliance

“The next generation of data systems won’t be defined by how much data they store, but by how intelligently they can turn it into insight,” said Yaroslav Tkachenko, founder of Irontools and data-streaming advocate. “Platforms like DeltaStream Fusion represent the future of streaming — where real-time, batch, and analytics converge into one fluid experience. As AI becomes more and more important in every data workflow, unifying these layers is key to unlocking truly real-time intelligence for modern enterprises.”

“Enterprises across verticals are struggling to stream data, but complexity and cost are holding them back from getting actionable, real-time insights,” said Hojjat Jafarpour, founder and CEO, DeltaStream, Inc. “DeltaStream is tackling this problem head-on with its unified data streaming platform. DeltaStream Fusion eliminates tool sprawl, reduces operating costs and enables every team to go from raw data to action in seconds.” 

DeltaStream Fusion is available now. Contact [email protected] or go to AWS Marketplace. Customers can deploy DeltaStream Fusion instantly — no infrastructure setup required — and begin building streaming, batch and real-time pipelines directly in SQL.

About DeltaStream

DeltaStream is a serverless streaming data platform company that empowers organizations to unlock real-time intelligence from their data. With DeltaStream Fusion, teams can unify streaming, real-time, and batch workflows within a single, easy-to-use platform—eliminating tool sprawl, cutting costs, and enabling faster data-to-action insights. Fusion integrates with popular streaming storage systems and features a powerful SQL interface for seamless data processing.

Press contact
Sammy Totah
[email protected]

DeltaStream, Inc., the serverless stream processing platform, today announced it has raised $15M in Series A funding from New Enterprise Associates (NEA), Galaxy Interactive and Sanabil Investments. The funding will accelerate the company’s vision of enabling a complete serverless stream processing platform to manage, secure and process all streaming data, irrespective of data source or pipeline.

Event streaming platforms such as Apache Kafka have become essential in today’s data-driven industries, with many sectors adopting real-time data streaming. As AI  advances, real-time data is becoming even more critical for  applications.

DeltaStream, founded by Hojjat Jafarpour, CEO and creator of ksqlDB, helps  organizations build real-time streaming applications and pipelines with SQL in minutes. The  platform leverages the power of Apache Flink© while simplifying its complexity, making it accessible to businesses of all sizes–from startups to large enterprises. DeltaStream is available as a fully managed service or a bring-your-own-cloud (BYOC) option.

“Streaming data is essential for modern apps, but it’s been difficult and costly to get value from it. DeltaStream’s serverless platform simplifies infrastructure management, allowing users to focus on building their applications,” said Hojjat Jafarpour. “To fully benefit from real-time data, customers need an organized view across all data stores, role-based access control, and secure sharing of real-time data. What Databricks and Snowflake did for stored data, DeltaStream does for streaming data.”

DeltaStream also integrates with Databricks and Snowflake, letting customers create real-time pipelines that quickly move data from streaming platforms like Apache Kafka to these systems.

“We’re seeing rapid adoption of streaming data and the rise of platforms such as Apache Flink©,” NEA partner and DeltaStream board member Aaron Jacobson explained. “However, one of the main challenges of using such systems has been their operational complexity. With DeltaStream, users have the power of Apache Flink© without having to deal with its complexity resulting in significantly cost effective and accelerated time to market for real-time data applications.”

“DeltaStream is leading the charge in real-time streaming, where speed, low latency and intelligent decision-making are critical for businesses to maintain a competitive edge.” said Jeff Brown, Galaxy Interactive Partner, “Their enterprise grade, secure, and scalable solution simplifies complex stream processing, allowing teams to focus on deriving insights rather than managing infrastructure.”

About DeltaStream

DeltaStream’s innovative stream processing platform harnesses the power of Apache Flink© to simplify and easily process real-time data. Furthermore, the platform provides governance, organization and secure sharing capabilities for streaming data across all streaming storage platforms including Apache Kafka, Apache Pulsar AWS Kinesis and many more. In addition to its SaaS offering, DeltaStream’s platform is also available in private SaaS(also known as Bring Your Own Cloud) deployment to address the needs of regulated industries with high data privacy and security requirements. DeltaStream’s platform seamlessly integrates with both Databricks and Snowflake platforms enabling customers to build real-time data pipelines to make data available in Databricks and Snowflake seconds after it is available in Streaming platforms such as Apache Kafka.
https://www.deltastream.io/

DeltaStream is exhibiting this week at the Current Conference in Austin, Texas.