Normalized claim
People served: 10,000 users increase
“Over 10,000 people have used Viktor since its launch.”
Bevar Ukraine — Agentic AI chatbot for refugee integration on AWS
Use case typeCustomer service agentUpdated Jun 21, 2025
Bevar Ukraine, a Danish nonprofit, built an AI virtual assistant to help Ukrainian refugees navigate daily life and social services in Denmark. The solution uses Amazon Bedrock and evolved from a RAG chatbot into an agentic system with about 150 specialized AI agents for translation, legal, healthcare, and orchestration tasks.
- Organization
- Bevar Ukraine
- Industry
- Other
- Location
- Denmark
- Published
- June 2025
Reported outcomes
100,000 questions
questions answered in first yearOther quantified impact
10,000 userspeople served4-5 usersmonthly active users
Strategic outcomes
Customer experience & trustProvided 24/7 immediate support for refugeesScale & capacityFreed staff and municipality resources from repetitive questionsRisk & complianceValidated compliance with European regulations before rollout
Why do we believe this?Outcome claims, sources, and evidence checks
Normalized claim
Monthly active users: 4-5 users increase
“This includes 4,000–5,000 monthly active users”
Normalized claim
Questions answered in first year: 100,000 questions increase
“The chatbot has answered over 100,000 questions in the first year”
Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks
- Customer
- Bevar Ukraine
- Provider
- AWS
- Maturity
- Production
- Linked source
- AWS Customer Stories
It was tested for months, checked for European regulatory compliance, and rolled out in more than 100 languages
Customer identity supportedSource describes one deploymentMaturity supported
Primary read
Use case focus
Showing 1 of 1
- 1Customer service agent
- The nonprofit's social and legal assistance center was overwhelmed by repetitive questions from refugees.
- A single employee could not keep up with questions about housing, schools, jobs, taxes, medical appointments, and language courses.
- The team created Viktor, a virtual assistant powered by Amazon Bedrock and a knowledge base of more than 400 official Danish resources.
- The system uses around 150 specialized agents and multi-agent orchestration to translate, route, and answer questions more precisely.
- It was tested for months, checked for European regulatory compliance, and rolled out in more than 100 languages.
- Over 10,000 people have used Viktor since launch.
- The chatbot supports 4,000 to 5,000 monthly active users and has answered 100,000+ questions in its first year.
- It provides 24/7 support and reduces waiting times that previously took weeks for simple questions.
Architecture
Viktor evolved from a single-LLM RAG architecture into a complex agentic AI system on AWS using Amazon Bedrock and around 150 specialized agents, including translation, legal, healthcare, and orchestrator agents. The team also followed the AWS Well-Architected Framework and validated compliance with European regulations before rollout.
Sources & evidence1
Evidence: Medium65/100Evidence strength
- Customer explicitly identified
- Deployment status explicitly supported
- Primary source available
- Quantified outcome available
- Technical implementation details available
Type: Customer StoryPublished: Jun 21, 2025Publisher: AWSEvidence: PrimaryConfidence: High
AI-generated summary. Verify important details with the linked sources before relying on this case.
Explore related AI use cases
Customer evidence
Provider evidence
Market context
AI patterns
Browse the catalog
Was this useful?
Community
Comments
No published comments yet.