ToolsSeptember 8, 2026via AWS Machine Learning Blog
How DiDi built intelligent contact center QA with Amazon Bedrock
Why it matters
A concrete deployment case study showing how enterprises are building custom AI workflows on foundation models to replace legacy vendor tools, with measurable wins in accuracy and speed.
Key signals
- DiDi built contact center QA system on Amazon Bedrock
- Intent verification accuracy improved from 38% to 86%
- Compliance scoring reached 90%
- Voice of Customer trend analysis reduced from hours to minutes
- System handles Spanish and Portuguese support
- Replaced third-party opaque tool with self-owned transparent system
- Intent verification accuracy: 38% → 86% (48-point lift)
- Compliance scoring: topped 90%
- Voice of Customer analysis: hours → minutes
- Languages: Spanish and Portuguese support
- Platform: Amazon Bedrock (managed LLM API)
- Use case: Contact center QA automation replacing third-party vendor
The hook
DiDi replaced opaque third-party QA with Bedrock—intent accuracy jumped from 38% to 86%, compliance scoring hit 90%.
DiDi built a transparent, self-owned contact center quality assurance (QA) system on Amazon Bedrock, replacing an opaque third-party tool. Intent verification accuracy rose from 38% to 86%, compliance scoring topped 90%, and Voice of Customer trend analysis dropped from hours to minutes across Spani…