Success Case | Transforming Information Access with Generative AI
Multinational insurer of assistance to individuals and professionals, with a portfolio of +30M clients in 49 countries and a network of +35,000 employees.
Quick view
The Challenge
Our client struggled with slow, manual, and inconsistent access to the general conditions of their policies, which were stored in an unstructured format. This led to operational inefficiencies and delayed or inaccurate responses to customers and agents, with inquiries taking between 4 and 7 minutes. The manual workflow created bottlenecks, increased costs, and introduced compliance risks in handling sensitive data. The company needed to modernize this process to maintain its agility and competitive leadership.
The Solution
Keepler partnered with this client to productionize their GenAI prototype. A robust and secure cloud-native architecture was designed in AWS, migrating from an on-premises environment to scalable enterprise services.
The solution uses Amazon Bedrock to generate intelligent, contextualized responses. For information search and retrieval (RAG), Amazon Kendra was implemented, enabling precise vector-based search supported by metadata. Policy documents are automatically synchronized each night from SharePoint to a data store in Amazon S3, ensuring data freshness.
A solid MLOps security framework was implemented, using Azure Active Directory (AD) for user authentication and AWS KMS for data encryption. Finally, the platform includes advanced monitoring (such as Langfuse) for full observability and traceability of interactions, providing employees with a secure chat interface for instant access to documentation.
Benefits
The implementation of this solution delivered an immediate and measurable business impact.
It improved operational efficiency, response consistency and accuracy (and therefore service quality and regulatory confidence), empowered agents and operations staff with greater autonomy (through instant access to verified information), and provided a scalable and secure foundation for future Generative AI innovations within the company.
- Up to 85% reduction in average response time per inquiry (from 4–7 minutes to under 30 seconds).
- Increase in autonomous inquiry resolution rate from 50% to 87% (without human escalation).
- 74% increase in self-service capability.
Keepler is a full-stack analytics services company specialized in the design, construction, deployment and operation of advanced public cloud analytics custom-made solutions. We bring to the market the Data Product concept, which is a fully automated, public cloud services-based, tailored software that adds advanced analytics, data engineering, massive data processing, and monitoring features. In addition, we help our customers transition to using public cloud services securely and improve data governance to make the organization more data-centric.
Would you like to talk about your business?
We can help you leverage the power of data to enhance your operations.