How Qubtrix automated customer support with an AI-powered chatbot built on AWS
About the Client
QUBTRIX is a Polish brand that creates innovative wooden construction blocks compatible with popular systems like LEGO®. Their products combine education, sustainability, and aesthetics, supporting the creative development of both children and adults. The company emphasizes local production, high quality, and functionality – offering solutions that inspire imaginative play and hands-on experimentation.
QUBTRIX operates a multichannel sales model – its products are available through its own e-commerce store, online marketplaces, industry trade fairs, and a physical showroom at Kraków’s Forum Designu. The brand’s dynamic growth and diverse customer base generate a high volume of inquiries related to product features, compatibility, orders, and availability.
The challenge
The QUBTRIX team was looking for a way to automate responses to frequently asked customer questions – without involving the support team. The challenge was to build an effective system capable of answering inquiries based on the company’s existing documentation and internal data. A key objective was to evaluate the effectiveness of the Retrieval-Augmented Generation (RAG) approach while leveraging AWS infrastructure to enable fast deployment, low maintenance costs, and scalability for future development.
Project objectives
- Automating customer inquiry responses using existing company materials — without requiring the team’s involvement in every interaction.
- Validating the effectiveness of the RAG (Retrieval-Augmented Generation) approach in the context of chatbots powered by large language models.
- Leveraging AWS services to build a scalable and flexible architecture with the lowest possible maintenance costs.
- Streamlining knowledge base updates by automating the process of ingesting new documents—eliminating the need for manual input. 1/2
- Collecting data from user interactions to continuously improve chatbot performance and gain deeper insights into user needs.
Pre-implementation situation
Before implementing the chatbot, Qubtrix faced several challenges that impacted both customer service efficiency and team productivity:
Employees were overwhelmed with handling repetitive inquiries, limiting their availability to focus on key tasks.
There was no automated system for updating information, which increased the risk of providing outdated data.
Customer support was primarily manual, restricting the speed and availability of assistance, especially outside of working hours.
It was essential to implement a tool enabling 24/7 efficient communication with customers and easing the team’s workload.
Our solution
To tackle Qubtrix’s challenges, we implemented an advanced chatbot powered by Amazon Lex technology, seamlessly integrated with AWS services to ensure stability, security, and scalability.
The chatbot features an automatic knowledge base update mechanism through integration with Amazon S3, allowing real-time enrichment and refreshing of information available to users without any manual intervention.
Additionally, we set up logging of all interactions in AWS CloudWatch, enabling thorough analysis of user behavior and faster detection and resolution of potential issues.
A key improvement was eliminating the “cold start” problem — the chatbot responds instantly, even after prolonged inactivity, providing a smooth and comfortable user experience.
Thanks to this solution, Qubtrix can efficiently handle customer inquiries around the clock, lighten the team’s workload, and quickly adapt to evolving business needs.
Tools and services used
In implementing the chatbot project for Qubtrix, we leveraged modern cloud solutions and AWS technologies to ensure reliability, scalability, and efficient system performance:
- Amazon Lex – as the primary platform for building and managing the chatbot’s conversational interactions,
- Amazon S3 – used for storing documents and files that serve as the chatbot’s knowledge base, enabling automatic updates of the knowledge repository,
- AWS Lambda – enables scalable, serverless processing of queries and handling of the chatbot’s business logic,
- AWS CloudWatch – used for monitoring and logging user interactions with the chatbot, enabling analysis and optimization of system performance,
- Amazon Bedrock – supports the language model used to generate responses to user queries,
- OpenSearch Serverless – creates and manages a dynamic knowledge base, enabling the chatbot to quickly retrieve the needed information.
Thanks to the combination of these services, we successfully built an advanced, secure, and easy-to-maintain system that meets Qubtrix’s needs.
Benefits for the Client
The chatbot implementation brought the company the following benefits:
Faster access to information – employees find the data they need more quickly, boosting their productivity.
24/7 customer support availability – the chatbot responds around the clock, improving customer satisfaction and engagement.
Automatic knowledge base updates – ensuring the chatbot always uses the latest information without manual intervention.
Full control and interaction analysis – conversation logs enable monitoring quality and continuously improving the chatbot’s performance.
Low maintenance costs – the optimized architecture provides a cost-effective solution without compromising quality.
Summary
The AI chatbot implementation at Qubtrix exemplifies a modern approach to automating customer service and streamlining business processes. By leveraging AWS technology and a well-thought-out integration, the client gained a tool that not only boosts team efficiency but also enhances customer satisfaction through 24/7 support. Cost optimization and automated knowledge base updates make the solution scalable and future-proof. This investment translates into real savings in time and resources, allowing Qubtrix to grow dynamically in the market.
