What changed
IskayPet, a leading pet care company in Iberia, implemented Amazon Bedrock Managed Knowledge Bases to unify corporate knowledge across multiple languages and countries. The solution allows employees to access operational information, such as return policies, veterinary protocols, and onboarding steps, without managing Retrieval Augmented Generation (RAG) infrastructure.
What this means in practice
The company’s solution answers natural-language questions grounded in their own corporate documents, supporting four languages and operations across three countries. It routes questions to the appropriate body of knowledge without cross-contamination and scales to new areas and countries without re-architecting. The system also cites sources so employees can verify answers and reuses existing SharePoint repositories without costly migration.
Limitations and availability
The solution was built using Kiro with Claude Haiku 4.5 on Amazon Bedrock. IskayPet plans to evaluate and adopt new models as they become available on Amazon Bedrock. The SharePoint connector enables advanced indexing, allowing content from images embedded in documents to be searchable without a separate preprocessing pipeline. The solution is available in AWS regions where Amazon Bedrock models are supported, and the exact availability depends on the region's supported models.
