DiDi Built an Intelligent Contact Center Quality Assurance System with Amazon Bedrock

Serdar HocamAuthor & Editor

Aiming to enhance the customer experience in its international operations, the company transitioned to a transparent and proprietary AI architecture instead of third-party solutions.

◉ 0 views
How DiDi built intelligent contact center QA with Amazon Bedrock | Amazon Web Services

The DiDi International Business Group has developed a new AI-powered quality assurance system for its contact center operations using the Amazon Bedrock infrastructure.

Transition to a New AI Architecture

DiDi's Customer Experience department moved away from previously used opaque third-party solutions, transitioning to a completely transparent and proprietary structure.

Multilingual and Versatile Scope

This newly developed system covers three distinct business lines—ride-hailing, food delivery, and financial services—along with Spanish and Portuguese languages.

Core Processing Pipelines

The system consists of three main processing pipelines: intent validation, compliance evaluation, and Voice of the Customer analysis.

Performance and Accuracy Improvements

In production validation testing, intent validation accuracy increased from 38 percent to 86 percent.

Efficiency and Speed Gains

While compliance scoring accuracy exceeded 90 percent, Voice of the Customer analysis reduced hours of manual summarization processes down to minutes.