Confluent has introduced Streaming Agents, a new capability within Confluent Cloud for Apache Flink, designed to help enterprises build and scale artificial intelligence agents that operate on real-time data. The launch aims to remove technical barriers for companies seeking to adopt agentic AI by unifying data processing with AI workflows and providing secure connections to large language models, tools, and business systems.
Tackling the challenge of scaling agentic AI
Many businesses exploring AI have struggled to move beyond pilot projects. According to research from IDC, organisations ran an average of 23 generative AI proofs of concept between 2023 and 2024, yet only three reached production, and just 62 per cent of those met expectations. The difficulty often lies in integrating real-time data into AI initiatives, leading to unreliable results and high costs.
Shaun Clowes, Chief Product Officer at Confluent, said, “Agentic AI is on every organisation’s roadmap. But most companies are stuck in prototype purgatory, falling behind as others race toward measurable outcomes. Even your smartest AI agents are flying blind if they don’t have fresh business context. Streaming Agents simplifies the messy work of integrating the tools and data that create real intelligence, giving organisations a solid foundation to deploy AI agents that drive meaningful change across the business.”
IDC’s Stewart Bond added that enterprises must prioritise solutions capable of secure integration with real-time data to provide the context needed for intelligent action.
Real-time decision-making with streaming data
Streaming Agents brings agentic AI directly into stream processing pipelines through Apache Kafka and Apache Flink. By combining data processing and AI reasoning, agents can adapt quickly to changing conditions, communicate with other systems, and act on real-time insights.
The technology is designed to operate continuously on behalf of businesses, processing high volumes of data and responding instantly to signals with context-aware reasoning similar to human operators. For example, a retailer could use Streaming Agents to monitor competitor pricing across e-commerce sites and automatically adjust its own prices to remain competitive.
Features and availability
Key functions of Streaming Agents include tool calling for context-aware automation using the Model Context Protocol, secure integrations with databases and APIs, enrichment of streaming data with external sources to improve decision accuracy, and replayability for safer testing and faster iteration.
These features aim to reduce complexity and cost while enabling teams to deploy production-ready AI agents more efficiently. Streaming Agents are now available in open preview, allowing enterprises to test and implement the technology.
Confluent said the new capability could help businesses move past stalled AI projects and unlock new business opportunities built on real-time intelligence.
