Duco Upgrades Financial Data Platform to Amazon Bedrock AgentCore
Financial operations platform Duco has rebuilt its AI stack using Amazon Bedrock and AgentCore to improve transaction processing and data isolation for major banks.
🔌Chips & HardwarePhoto: AWS Machine Learning Blog
Key points
- Duco processes 20 billion monthly transactions for over 200 financial clients.
- The platform migrated from classical machine learning to Amazon Bedrock and AgentCore.
- Strict multi-tenant isolation ensures separate data contexts for each bank.
Financial operations platform Duco has overhauled its artificial intelligence infrastructure, migrating to Amazon Bedrock and Amazon Bedrock AgentCore to handle large-scale data automation for tier-one banks and asset managers. The company currently manages 20 billion transactions every month for over 200 clients, including seven of the top twenty global banks. By upgrading its systems, Duco aims to accelerate deployment times and lower onboarding costs while maintaining strict data separation between different financial institutions.
Previously, Duco relied on traditional machine learning models for predictive labeling and pattern recognition to automate more than 20,000 workflows. According to Chief Architect and Innovation Officer Rob Gates, those classical models lacked the agility required to handle changing business rules and novel scenarios without extensive historical training data. To overcome these limitations, the company adopted Amazon Bedrock for managed access to foundation models and AgentCore to scale its autonomous agents.
The updated architecture integrates closely with Duco's existing setup using LangChain and LangGraph. It enforces strict multi-tenant isolation through identity authentication and infrastructure controls, ensuring that each client's data and context remain completely separate. Furthermore, the platform employs a multi-model strategy, assigning specific tasks to the most suitable foundation model to optimize both performance and cost efficiency.
What to watch next: Monitor how financial technology providers adopt multi-agent architectures for regulatory compliance and secure enterprise data handling.
Read next
🧠AI AnalysisNvidia's Interim CEO Looks to Calm Waters
Nvidia's interim CEO Colette Kress is working to stabilize the company after its former leader's resignation, using a new catchphrase to reassure employees and investors.…
⚖️Regulation & PolicyAirbnb optimizes coding agent performance using Amazon Bedrock metrics
Airbnb has implemented a cost-per-task evaluation framework to manage the performance and expenses of its internal coding agents hosted on Amazon Bedrock.
🧠AI AnalysisAWS Outlines Responsible AI Framework for Database-Integrated Chatbots
Amazon Web Services has released a technical blueprint for integrating responsible AI controls directly into Amazon Aurora PostgreSQL databases to enhance chatbot securit…
🏢Big TechAnduril CEO Palmer Luckey Calls for Jensen Huang’s Reinstatement at Nvidia
Anduril CEO Palmer Luckey and other industry leaders are pushing for the return of Jensen Huang to Nvidia, arguing his departure weakens U.S. interests. Meanwhile, Nvidia…
🌐Work & SocietyDC Studios pivots strategy under Skydance leadership
Following Skydance’s acquisition of Warner Bros. Discovery, DC Studios is shifting its cinematic strategy toward high-profile heroes, potentially drawing inspiration from…
⚖️Regulation & PolicyRe-evaluating Galera Node Weights: The Risks of Custom Quorum Configurations
Database administrators are being cautioned against using custom weights to manage even-numbered Galera clusters, as the practice introduces significant risks of instabil…