Thunai Overhauls Customer Support with IBM Confluent Real-Time Data Streaming

Thunai, a company that specializes in automating customer support using AI agents, has found a solution to its data challenges by partnering with IBM Confluent. The firm's agentic AI platform relies on real-time data streaming to keep its AI agents up-to-date and accurately interpret user intent. However, before adopting Confluent, Thunai faced numerous business and technical challenges that hindered the effectiveness of its customer support system.

One major issue was the reliance on batch processing, which led to slow decision-making and context switching fatigue among customer service agents. Additionally, data was spread across multiple systems, making it difficult for Thunai's AI platform to access accurate and real-time information about customers.

The technical challenges were equally daunting, with legacy integration approaches and traditional ETL tools failing to meet the firm's low-latency streaming requirements. Moreover, Thunai's knowledge store was unable to stay continuously updated, resulting in stale context for its AI agents.

With Confluent's help, Thunai has transformed its data architecture into a real-time, event-driven system that can process high-throughput workloads and provide accurate insights for customer support. The new platform allows Thunai's agentic AI to continuously update its knowledge base with fresh context from live customer conversations, CRM events, user actions, and other connected applications.

The benefits of this new architecture are numerous, including increased efficiency, improved accuracy, and enhanced customer satisfaction. With Confluent Cloud, Thunai has been able to scale its system to meet the demands of high-traffic periods, ensuring that its AI agents can safely act on behalf of users and support teams in real-time.