How MRH Trowe enabled secure self-service AI agents in financial services

This article describes how MRH Trowe, a German insurance broker, enabled secure self-service AI agents for approximately 400 employees while maintaining compliance with financial sector regulations.

• Combined Strands Agents, Amazon Bedrock AgentCore, and LibreChat to balance self-service access with security and governance
• First production agent automates meeting minutes generation from Microsoft Teams transcripts with user identity isolation
• Deployed in AWS Europe (Frankfurt) Region with private connectivity, session isolation, and centralized governance
• Infrastructure and token costs approximately $14 per seat per month with 40% reduction potential through optimization
• Supports multi-step workflows, institutional data integration, and contextual responses grounded in internal systems
• Leverages Active Directory SSO, token budgets, and role-based access controls for regulated compliance

MRH Trowe demonstrates how regulated financial services organizations can provide broad AI access while maintaining security, cost transparency, and compliance requirements through a centrally governed platform.