Case Study: Tractable achieves scalable annotation quality control with Encord

Tractable trains 40+ annotators with Cord's QA and review tools

Tractable, an AI company that provides visual damage assessment tools for insurance providers, faced challenges in scaling its data annotation projects. As their annotation tasks grew more complex, involving detailed segmentation and intricate ontologies, their existing platforms lacked the necessary quality assurance and annotator review functionality. This created bottlenecks and risks for their rapidly growing remote annotation team. To address this, they turned to the vendor Cord for a more capable training data platform.

Cord provided a user-friendly platform with robust quality assurance and annotator monitoring features. This solution enabled Tractable to efficiently train and manage a team of over 40 annotators while maintaining strong quality control. Cord's API integration with S3 servers also ensured compliance with global data governance rules. The result was a more efficient annotation workflow that balanced speed and quality, allowing Tractable to continue building sophisticated AI models for insurance.

Tractable