From portal-hopping to instant answers: HEMA’s journey with MCP and Amazon Bedrock
Not a pilot. HEMA deployed an internal AI agent across developer teams using MCP and Bedrock.
Why it matters
A concrete enterprise agent deployment case study showing how MCP + Bedrock AgentCore solves the real problem of fragmented developer tools — relevant to practitioners building or deploying agents in large orgs.
The key facts
12 to knowHEMA (100-year-old Dutch retailer) built HAL, an internal AI assistant
Built on Amazon Bedrock AgentCore
Uses Model Context Protocol (MCP) for tool integration
No AWS credentials on client; security anchored in Microsoft Entra ID
Deployed across developer teams (multi-tool consolidation use case)
Solves 'portal-hopping' problem — delivering answers in existing developer workflows
HEMA built HAL, an internal AI agent on Amazon Bedrock AgentCore
Uses Model Context Protocol (MCP) for knowledge governance
Integrates with Microsoft Entra ID for identity/security anchoring
Eliminates AWS credentials on client side
Agent deployed across developer portal workflows (reduces multi-system context-switching)
100-year-old Dutch retailer signals enterprise adoption maturity
Go to the source
AWS Machine Learning Blogaws.amazon.com
Publisher excerpt: HEMA, a 100-year-old Dutch retailer, turned developer portal-hopping into instant answers by building HAL, an internal AI assistant on Amazon Bedrock AgentCore. Using Model Context Protocol (MCP), HAL delivers governed knowledge inside the tools teams already use, with no AWS credentials on the…