This course was developed by members of AWS Technical Field Communities (TFC), an AWS community of technical experts. The content is intended to complement our standard training curriculum and augment your AWS learning journey. We are aware some courses have accessibility limitations and are working to address. If you require accommodation, please contact AWS Training and Certification Customer Support.
Learners will discover how to build a cost-effective two-model pipeline on Amazon Bedrock for digitizing scanned documents at scale. The course covers pairing Amazon Nova 2 Lite -- for native multimodal extraction including photo detection, name extraction, and metadata -- with Claude Sonnet 4.6 for spatial reasoning that matches names to faces based on page layout. Participants will learn how this approach achieved 93 percent high-confidence associations across 336 yearbook pages while costing about two-thirds less per page than a single-model alternative.
