LoanPro deploys a multi-agent AI assistant for loan servicing on the Amazon Bedrock platform
LoanPro, together with AWS and partner AllCloud, built a multi-agent AI assistant on the Amazon Bedrock platform using Claude models. According to the published case study, it reduces service agents' call times by up to 15%.
LoanPro, a company operating a loan management platform used, according to its own figures, by more than 450 lenders, developed a multi-agent AI assistant for call center service agents in collaboration with the AWS Generative AI Innovation Center and partner AllCloud. During a call with a client, the assistant automatically provides agents with synthesized account context and recommendations for next steps, instead of the agent having to manually piece together information from dozens of fields, transaction history, and task queues.
The system runs on the Claude Sonnet 5.5, Claude Haiku 5.5, and Claude Opus 5.5 models from Anthropic on the Amazon Bedrock platform, with orchestration handled by Amazon Bedrock AgentCore, and agent logic built on the open-source Strands Agents SDK framework. The architecture is split into specialized agents — one handles data retrieval and account summarization, the other handles action selection and recommending next steps — coordinated by a so-called composer agent, which routes requests to the appropriate specialist.
According to a case study published by AllCloud, deploying the assistant led to call times being shortened by up to 15%. The article further describes technical details of request processing (API Gateway, Lambda, SQS, MCP server), but these are not essential for a business summary. The rest of the case study text was not available in the source.
Why it matters
This case shows a concrete, measurable benefit of a multi-agent AI architecture in real-world call center operations — shortened call time is directly tied to lower operating costs and faster client service. For companies in financial services and loan management, it serves as a reference implementation for how to deploy specialized AI agents on top of an existing system without needing to build custom infrastructure for running the models.
Relevant practical impact
What this means
For a business
A multi-agent AI assistant built on top of data from an existing loan management system can, according to a documented case study, reduce customer handling time by up to 15%, which lowers call center operating costs and speeds up request processing without the need to build custom AI infrastructure from scratch.
ProductivityCheck the original
Event sources
only one source so far · 1 publisher, 0 independent. We count feeds from the same owner only once.