A UK challenger bank

FintechChallenger Bank

Cloud-native credit decisioning for a digital-first bank

From blank sheet to production-grade credit decisioning in four months.

Platform & ArchitectureAI & DecisioningCloud & Payments
→

Designed and delivered ~20 microservices credit decisioning platform in 4 months

→

Event-driven, cloud-native architecture on public cloud

→

Single platform supporting multiple product types (overdrafts, loans, etc.)

→

Enabled rapid experimentation with lending journeys and risk strategies

Request a detailed version of this case study →

Engineering Facets

Domain AlignmentCloud NativeTestabilityModularityObservabilityIncremental Delivery

Expertise

Greenfield Product EngineeringDomain-Driven Design PracticeEvent-Driven Architectures

Accelerators

Domain-Driven DesignCloud Native FoundationPlatform Core ScaffoldService Pattern LibraryAI-Native Engineering PipelineGenAI Evaluation Harness

Capabilities

Platform Engineering
“Building the right internal platform creates an enduring asset. Buying a monolithic decisioning engine trades speed for future constraints.”

Context

The bank was pursuing a full licence and needed a modern lending stack that could support their first credit products and evolve with their roadmap. Buying a monolithic decisioning engine would trade speed for future constraints; building the right internal platform would create an asset. They chose to work with a small, senior Bugni Labs squad to co-create that platform.

The Challenge

  • →

    Design a credit decision runtime that could support multiple products and rapidly changing policies

  • →

    Achieve production readiness under tight regulatory and delivery timelines

  • →

    Maintain explainability across affordability, eligibility, credit scoring and limits

  • →

    Blend internal models, third-party scores and policy decisions into coherent journeys

Our Approach

Domain-driven design and service decomposition

Clear bounded contexts: applications, products, scoring, limits, decisions, fulfilment. Each context with its own models and APIs.

Event-driven microservices on cloud

Asynchronous processing for heavy or long-running work. Horizontally scalable stateless services for peak periods.

Business workflows with BPMN 2.0

Lenders could design product journeys (e.g. overdrafts vs. personal loans) with re-usable building blocks.

Observability and business metrics

End-to-end instrumentation for application drop-offs, decision times, policy overrides and exceptions.

Engagement Timeline

Domain Modelling

3 weeks

Defined bounded contexts for applications, products, scoring, limits, decisions and fulfilment with clear API contracts.

Platform Build

10 weeks

Built ~20 microservices with event-driven processing, BPMN workflows for product journeys, and end-to-end instrumentation.

Production Launch

3 weeks

Hardened for production, launched first lending products (overdrafts), and validated regulatory compliance and performance.

Outcomes

  • →

    Production-ready platform delivering the bank's first lending products within a four-month window.

  • →

    Single decisioning foundation for multiple product lines (overdrafts, personal loans, future products).

  • →

    The bank gained an internal asset that can adapt to changing risk appetite, regulation and growth.

Technologies & Patterns

Cloud-native microservicesBPMN 2.0Event-driven architectureDomain-driven designPublic cloud (AWS)Observability stack

What Our Client Said

“The engineering quality and clarity gave our auditors confidence from day one.”

Credit Decisioning Platform engagement

“The speed of delivery improved because the complexity became clear.”

Credit Decisioning Platform engagement

Related

Perspectives that argue this

Guides that cover this

Want the full story?

Client names, detailed architecture diagrams, metrics, and implementation insights are available on request. We're happy to walk you through the engagement.

Discuss a similar challenge