Confluent Cloud Launches Streaming Agents for AI

News Desk - 21/08/2025

Confluent, Inc. (Nasdaq: CFLT), the data streaming pioneer, announced Streaming Agents, a new feature in Confluent Cloud for Apache Flink®. The capability allows businesses to easily build and scale AI agents that monitor, reason, and act on real-time data.

Streaming Agents aim to remove barriers to enterprise-grade agentic AI. They unify data processing and AI workflows while providing secure connections to all business systems, including large language models (LLMs) and embedding models. This accelerates AI adoption, improves workflow efficiency, and creates new business opportunities.

Shaun Clowes, Chief Product Officer at Confluent, said, “Streaming Agents simplifies the messy work of integrating tools and data. Organizations can now deploy AI agents that drive meaningful business change.”

IDC research shows companies ran an average of 23 generative AI proofs of concept between 2023 and 2024. However, only three reached production, with just 62% meeting expectations. Stewart Bond, VP of Data Intelligence at IDC, emphasized that businesses should adopt agentic AI solutions that integrate easily and leverage real-time data for smarter decisions.

Streaming Agents features include:

Tool calling for context-aware automation: Agents select the right external tools, such as databases, SaaS platforms, or APIs, to take meaningful action.
Secure connections for AI integration: Flink enables connections to models and databases while protecting sensitive credentials and centralizing management.
External tables and search for AI accuracy: Integrating relational databases and REST APIs improves AI decision-making, vector search, and RAG applications.
Replayability for safe iteration: Agents can be tested with real data without side effects, enabling faster iteration and A/B testing.

Streaming Agents operate continuously, processing high-volume data streams and responding instantly to real-time signals. Businesses can use them for competitive pricing, automated monitoring, and other AI-driven tasks.