Build a FinOps agent using Amazon Bedrock AgentCore
Machine Learning Blog
This article explains how to build a FinOps agent using Amazon Bedrock AgentCore to manage AWS costs across multiple accounts through natural language queries.
- Consolidates data from AWS Cost Explorer, Budgets, and Compute Optimizer into single interface
- Uses Amazon Bedrock AgentCore Runtime with Claude Sonnet 4.5 and Model Context Protocol servers
- Maintains 30-day conversation memory for contextual follow-up questions
- Provides 24 specialized cost management tools covering analysis and optimization
- Architecture includes authentication layer, MCP servers, Gateway, and main agent runtime
- Deployment uses AWS CDK with five CloudFormation stacks in us-east-1 region
- Frontend hosted on AWS Amplify with Amazon Cognito user authentication
- Supports queries like cost analysis, savings opportunities, and optimization recommendations
The solution enables finance teams to access cost data through conversational AI without manual console navigation, with deployment taking approximately 15-20 minutes.
The AWS News Feed is currently looking for gold sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.
Related articles
Apr 18
2025
2025
Build a FinOps agent using Amazon Bedrock with multi-agent capability and Amazon Nova as the foundation model
Mar 31
2026
2026
Build reliable AI agents with Amazon Bedrock AgentCore Evaluations
Jan 23
2026
2026
Build AI agents with Amazon Bedrock AgentCore using AWS CloudFormation
Oct 13
2025
2025
Make agents a reality with Amazon Bedrock AgentCore: Now generally available
The AWS News Feed is currently looking for silver sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.