Staff ML Engineer - Real-Time Fraud Platform Lead

MoonPay; Greater London; Hybrid; GBP 120,000 - 180,000; Full time

Benefits offered by this job

Equity package

Hybrid working schedule

Private healthcare benefits

Lunch credit

Home office setup allowance

Annual training budget

Job summary

MoonPay is hiring a Staff Machine Learning Engineer in London to own the real-time decisioning platform for transaction scoring. You will build, ship, and maintain the ML lifecycle, from data pipelines to in-flight scoring, with a focus on fraud detection and rapid iteration.

You will lead the design of feature infrastructure, ensure robustness under latency budgets, and scale the platform as volume and model complexity grow.

Qualifications

• Experience building and operating high-availability services with strict latency budgets.
• Holistic view of architecture, managing degradation and ensuring responsiveness when dependencies fail.
• Clean, well-tested code that is easy to inherit and extend.
• End-to-end ownership of feature or data pipelines, with metrics alignment.
• Ability to turn vague problems into defined solutions and lead others.

Responsibilities

• Own services that score transactions in-flight within a strict latency budget.
• Develop feature infrastructure across batch, near-real-time, and in-request paths.
• Ensure models behave in production as in offline evaluations.
• Scale platform with growing volume and model complexity.
• Mature tooling for safe model updates (replay, shadow, staged rollout).

Skills

Real-time serving

Systems thinking

Engineering craft

Pipelines in production

Ambiguity & influence

Tools

GCP

BigQuery

Bigtable

Memorystore

Vertex AI

Kubernetes

Job description

MoonPay is hiring a Staff Machine Learning Engineer in London to own the real-time decisioning platform for transaction scoring. You will build, ship, and maintain the ML lifecycle, from data pipelines to in-flight scoring, with a focus on fraud detection and rapid iteration.

You will lead the design of feature infrastructure, ensure robustness under latency budgets, and scale the platform as volume and model complexity grow.