Novo Nordisk modernizes manufacturing process documents with Amazon Bedrock AgentCore
Novo Nordisk built an intelligent support system for manufacturing process documents using Amazon Bedrock AgentCore and a GraphRAG architecture that combines a knowledge graph with hybrid search. The managed agent framework coordinates retrieval and response, and a guided mode clarifies ambiguous queries, reducing tasks that previously took hours or days to minutes. This enables faster, more accurate, and traceable access to regulated documentation for subject‑matter experts.
⚡ Key Takeaways
- The solution reduced document‑query work from hours/days to minutes.
- Architecture uses GraphRAG (knowledge graph + hybrid search) to retrieve contextually accurate information.
- Managed agent framework (AgentCore) orchestrates end‑to‑end retrieval and response, ensuring traceability.
- Guided mode prompts users to clarify ambiguous questions before initiating search, improving precision.
- Designed for regulated pharmaceutical environment, emphasizing accuracy and auditability.
Engineers can leverage Bedrock AgentCore and GraphRAG to build compliant, high‑accuracy AI assistants that dramatically cut manual lookup time while maintaining traceable, cited answers—critical for regulated industries.
✅ Practical Steps
- Deploy the managed agent framework (AgentCore) on AWS as described to coordinate retrieval and response.
- Implement a GraphRAG pipeline that integrates a knowledge graph with hybrid search for document retrieval.
- Enable guided mode in the agent to prompt users for clarification before executing searches.
Want the full story? Read the original article.
Read on AWS ML Blog ↗