Skip to main content
Case Study

How MarvelX is Scaling Insurance Processing at the Speed of AI with Weaviate

Summary

MarvelX AI agents, powered by Weaviate, handle high-volume, high-stakes insurance workflows, delivering rapid claims processing with better-than-human accuracy. This enables their customers to scale efficiently while maintaining control over data security.

Challenge

Insurance and related financial institutions struggle with manual, repetitive workflows that slow down service, impair the customer experience, and limit scalability. Many still rely on tools like spreadsheets and outdated software, resulting in process bottlenecks and human error, especially as scale increases. For these businesses, high accuracy, speed, and reliability are non-negotiable, as any slip in performance directly affects customer satisfaction and operational costs.

Why did MarvelX choose Weaviate?

MarvelX had the following requirements for a vector database:

  • Speed and scalability: Their system needed to process large volumes of claims in real time, with Weaviate’s speed significantly outperforming competitors.
  • Multi-tenant security: With roots in digital banking and strict enterprise requirements, trust in data integrity and security was critical.
  • Multimodal search: They needed to store and search across multiple data types–like policies, invoices, attachments–to support advanced similarity checks and contextual matching.
  • Support and partnership: Positive interactions and ongoing innovation with the Weaviate team supported MarvelX’s fast-moving product development.

“Before MarvelX, I implemented Weaviate in banking where security was really important. Being able to have trust in something battle tested is very important. We also benchmarked multiple options and Weaviate’s speed blew the others out of the water. It wasn’t even close. As both our workload and customer base grew, the benefits just compounded,” notes el Hassouni.

What solution did MarvelX build with Weaviate?

MarvelX AI uses Weaviate as a foundation for their autonomous agent platform, allowing them to deliver:

  • AI-driven claims automation: Deployed agents accurately handle claims end-to-end with accuracy comparable or superior to scaled human operations.
  • Superior speed: Weaviate allowed MarvelX to process claims in seconds.
  • Longtail coverage: Through iterative feedback loops with subject matter experts (SMEs), MarvelX’s AI agents address rare and nuanced use cases, allowing more of the workflow to be reliably automated over time.
  • Enterprise-grade security: Each client is hosted as a separate tenant, deployed on either their own or MarvelX’s cloud, ensuring rigorous data separation and compliance.

What will MarvelX do next?

MarvelX’s platform is evolving to deliver even more agentic workflows, enable advanced customization for customers, and further improve automation rates. As industry requirements shift, MarvelX relies on Weaviate’s rich feature set and responsive support to remain agile and innovative.

“Our approach is intentionally collaborative. By involving our customers in refining agent workflows, we not only build trust, but also help them generate their own unique AI Agents. Weaviate’s flexibility lets us tailor every deployment to evolving needs,” said el Hassouni.

Results

90%+ automation potential

Achieved via automated end-to-end claims processing.

99.9% faster turnaround

Compared to manual claims processing, which can take days to weeks, Weaviate helps MarvelX deliver accurate results in seconds.

Scaling client operations 10X

Clients can drastically increase their claims throughput with the same team size, having human specialists shift from manual entry to orchestration and quality control.

Trust and building customer IP

Customers’ SMEs participate in training and refining agents, building internal expertise for customer organizations.

The startup was founded by Ali el Hassouni, former Head of Data and AI at bunq and a PhD in reinforcement learning and generative AI. MarvelX focuses on solving operational inefficiencies in insurance and adjacent financial services industries. While demand to incorporate more efficient AI processes is high, most organizations struggle to take their ideas into production. After mapping out workflows and identifying large, underserved markets, the company set out to build AI agents specifically for high-volume insurance processes. The result is a precise and intentional use of AI to transition from repetitive, manual work to reliable, intelligent operations that scale.

Our team and community are here to support you at every stage of your AI journey.

Get Started with Weaviate

Please leave your contact details below and one of our sales representatives will reach out to you within 24 hours.