customer_support · finance · workflow

bunq handles 97% of user support with Amazon Bedrock multi-agent AI system

bunq's traditional support systems could not keep up with round-the-clock multilingual demand from 20 million users, creating bottlenecks and straining internal resources. As bunq scaled its initial multi-agent architecture, a router-based design became a single point of failure due to routing complexity, overlapping agent capabilities, and a scalability bottleneck that required comprehensive testing for every new agent added.

How it works
Common implementation structure
How this type of workflow is generally built, generalized across documented cases — not tied to any one vendor's stack. Click any stage to read what happens there. Specific products that implement these stages appear in “Tools commonly seen” below.
Stage 1 · User submits natural language request
Users interact with Finn through bunq's application and web interface using natural language for their requests.
Tools used
Amazon BedrockAmazon ECSAmazon OpenSearch ServerlessAmazon SageMakerAmazon S3Amazon MemoryDBAWS WAFAmazon CloudFrontAWS CloudTrailAmazon GuardDutyAmazon CloudWatch
Outcome

Finn now handles 97% of bunq's user support activity with average response times of 47 seconds, was deployed in 3 months from concept, and supports 38 languages, positioning bunq as Europe's first AI-powered bank.

What failed first

The initial router-based multi-agent architecture became a single point of failure: routing logic grew too complex, agents required access to the same data sources, and every new specialized agent required comprehensive testing of all routing scenarios.

Results
Time saved47 seconds
Volume97%
Running sinceJanuary 2025 (concept start; production within 3 months)
Source

https://aws.amazon.com/blogs/machine-learning/how-bunq-handles-97-of-support-with-amazon-bedrock?tag=soumet-20

How we source this →

Grounding & classification
Source type: technical build writeup
50 fields verified against source quotes, 1 dropped as unverifiable.
agentic workflowconversational aidocument aimulti agent workflowragspeech to textsummarizationsupport agenttranslationchat transcriptinvoiceknowledge basefailure mode describedmetric backednamed customerproduction runtime claimedtools describedvendor confirmedworkflow describedbankingautomation ratedeflection rateemployee productivityresponse time reductiontechnical build writeupcall center aicustomer supportautonomous resolutionescalation workflow