How Condé Nast built multimodal video discovery with Amazon Bedrock
250 minutes to 2 minutes. How Condé Nast cut video discovery time by 125x using multimodal search on Bedrock.
Why it matters
Concrete deployment case study showing multimodal retrieval reducing manual search overhead in large media libraries. Demonstrates practical ROI of Bedrock + OpenSearch integration for editorial workflows, though limited to a single use case without broader adoption metrics.
The key facts
15 to knowCondé Nast video library: 140,000+ videos
Time per discovery task: 250 minutes → under 2 minutes (99.2% reduction)
Built on Amazon Bedrock and Amazon OpenSearch Service
Partnered with AWS Generative AI Innovation Center
Editorial teams affected: Condé Nast's content discovery workflows
Multimodal retrieval: search by description + visual features, not titles alone
Published Sep 29, 2026 on AWS ML blog (vendor case study)
Condé Nast: 140,000+ videos in library
Before: 250 minutes average per search task (metadata-only)
After: under 2 minutes per task (multimodal)
Technology stack: Amazon Bedrock, Amazon OpenSearch Service
Partner: AWS Generative AI Innovation Center
Use case: editorial video discovery workflow
No pricing, compute cost, or model details disclosed
No independent measurement or comparison offered
Go to the source
AWS Machine Learning Blogaws.amazon.com
Publisher excerpt: Condé Nast's editorial teams spent an average of 250 minutes per task searching a library of more than 140,000 videos using only titles and descriptions. Working with the AWS Generative AI Innovation Center, they built a multimodal video discovery solution on Amazon Bedrock and Amazon OpenSearch…