What changed
Novo Nordisk, a leading pharmaceutical company, has deployed an intelligent support system powered by agentic AI on AWS to modernize its manufacturing process documentation. This system, built using Amazon Bedrock AgentCore, aims to reduce lead times, improve documentation visibility, and make critical manufacturing knowledge more accessible.
What this means in practice
The solution addresses the challenges of managing a large portfolio of controlled documents, including standard operating procedures, work instructions, forms, and training materials. By leveraging a GraphRAG architecture, the system combines a knowledge graph with hybrid search to retrieve the most relevant and contextually accurate information. AI agents, built on a managed agent framework, coordinate the retrieval and response process end-to-end. The guided mode helps clarify ambiguous questions before a search is initiated, making responses more precise and defensible.
Limitations and availability
The solution is currently being used by Novo Nordisk's Manufacturing Technology team within the Tablet area. It is built around a GraphRAG architecture, combining a knowledge graph in Amazon Neptune with a hybrid search index in Amazon OpenSearch Serverless. The system is designed to meet the high standards of accuracy and traceability required in a regulated pharmaceutical environment. The team used Amazon Bedrock AgentCore, a platform to build, connect, and optimize agents at scale, and the agent runtimes are built with the Strands Agents SDK and powered by Anthropic Claude models on Amazon Bedrock. The ingestion pipeline processes documents through Amazon Bed, orchestrated with AWS Step Functions and AWS Lambda, and the stack is defined as infrastructure as code with the AWS Cloud Development Kit (AWS CDK).
