How Duco transforms financial data automation with Amazon Bedrock AgentCore
Duco, a data automation platform for tier-one banks, rebuilt its agentic AI stack on Amazon Bedrock and Amazon Bedrock AgentCore. With strict per-tenant isolation, a multi-model strategy, and an MCP skills architecture that cut context overhead by 90%, Duco cut reconciliation deployment time 10x and onboarding costs 70%.
First seen by Oriel on .
AI summary
This AWS case study describes how Duco uses Amazon Bedrock and Amazon Bedrock AgentCore to scale isolated, multi-model financial-data agents while reducing context and onboarding costs.
- Duco processes 20 billion transactions monthly for more than 200 clients, including seven of the top 20 banks and ten of the top 20 asset managers.
- Each tenant receives isolated infrastructure, authentication, AgentCore runtime, Gateway, MCP servers, and session context, with user identity propagated through authorized tool calls.
- Duco retains its LangChain and LangGraph code while assigning Anthropic Claude and Amazon Nova 2 models to planning, reasoning, routing, and summarization.
- An MCP skills architecture reduced static tool context from approximately 200,000 to 20,000 tokens, preserving prompt caching and improving evaluated response quality.
- Agents deploy reconciliations 10× faster; one global bank reduced onboarding costs by 70% and increased straight-through processing by more than 20%.
#amazon bedrock #agentic ai #mcp #financial services
Generated by gpt-5.6-sol. Check the original for details.
Aliases
| Source | External ID | URL |
|---|---|---|
| AWS Machine Learning Blog | 4d56ba5eeb480a2a306a4f4e8ec1bce9830a8eef | https://aws.amazon.com/blogs/machine-learning/how-duco-transforms-financial-data-automation-with-amazon-bedrock-agentcore/ Read discussion |