인피니언 8월20일부터
This page was machine-translated and may differ from the original. View original

OutSystems Supports Personal Loan Business Automation with Financial Agentic Solution

Google 우선 소스Published2026.08.28 09:08


 
Launch of 'Agentic Loan Application' Integrating AI Agents, Workflows, and Customer App
 
As AI adoption in the financial sector transitions beyond the experimental stage to practical business application, OutSystems has expanded its Agentic Industry Solutions portfolio targeting the advancement of personal loan operations. The solution is characterized by integrating AI agents, customer applications, and predefined workflows under a single governance framework, enabling financial institutions to flexibly apply the system in accordance with their own policies and existing banking systems.
 
OutSystems unveiled the Agentic Loan Applications on the 28th.

The solution is built on the OutSystems Agentic Systems Platform and combines mobile and web applications, data models, governance-applied agents, and predefined workflows to automate manual work and document processing in the loan review process, the company announced.

KeyBank, Paragon Bank, and Axos Bank are reportedly utilizing the solution in regulatory decision-making processes.
 
Luis Blando, Chief Product and Technology Officer (CPTO) at OutSystems, stated, "What banks need is not individually operated agents, but agentic systems with governance applied that can improve actual customer journeys and meet risk review requirements. By integrating agents into auditable workflows, we enable financial institutions to apply AI to complex loan processes."
 
Agent Evaluation and Model Selection Framework
 
Agentic Loan Application performs document collection and verification, identity verification, and sanctions screening before compiling the loan application and transferring it to the bank's existing loan execution system.

Banking agents undergo pre-testing against dozens of criteria including relevance, accuracy, and personally identifiable information (PII) protection, and undergo revalidation procedures when models, prompts, or tools are changed, OutSystems explained.
 
Amazon Bedrock is utilized in model selection.

OutSystems pre-selected models that deliver optimal performance at costs up to 82% lower through cost-to-performance evaluation, and stated that banks can configure the system to match their products and policies while maintaining data control authority.

Developer components are also provided on OutSystems Forge through the OutSystems Banking Agents Kit.
본 기사에 대한 정정·반론·추후보도 청구는 보도 청구 안내를, 그간 게재된 보도문은 정정·반론보도 모아보기를 참고해 주세요.
명세환 기자
명세환 기자