Senior ML Ops Engineer: AI Deployment & Governance

Mastercard; Blanchardstown; On-site; EUR 110,000 - 150,000; Full time

Job summary

Mastercard is seeking a Senior Machine Learning Ops - AI Engineering to build and operate pipelines, deployment workflows, and production-readiness practices turning trained models into governed services. You will own experiment tracking and model registries (MLflow or equivalent), implement drift and performance monitoring, and enable safe release strategies for model versions.

Key responsibilities include orchestrating training and inference workloads on Databricks, embedding security into

Qualifications

• Experience with MLOps tooling and practices: experiment tracking, model registries, and safe deployment/rollout patterns (e.g., MLflow).
• Experience supporting AI/ML workloads: deployment pipelines, batch/streaming inference, and training vs serving operations.
• Strong CI/CD experience for production environments with appropriate tooling and patterns.

Responsibilities

• Own experiment tracking and model registry practices (MLflow or equivalent).
• Implement drift and model-performance monitoring for data and embeddings.
• Implement safe model release, canary/shadow deployments, and rollback procedures.
• Onboard Databricks workflows for training and inference jobs.
• Design observability: logs, metrics, tracing, SLIs/SLOs for real-time and batch workloads.
• Build CI/CD pipelines for AI/data workloads and embed security practices (secrets, least-privilege).
• Support incident response and post-incident improvements.

Skills

MLOps tooling

Experiment tracking

Model registry

CI/CD pipelines

Databricks

Docker

Kubernetes

Cloud platforms

Security in CI/CD

Observability

Tools

MLflow

Terraform

Job description

Mastercard is seeking a Senior Machine Learning Ops - AI Engineering to build and operate pipelines, deployment workflows, and production-readiness practices turning trained models into governed services. You will own experiment tracking and model registries (MLflow or equivalent), implement drift and performance monitoring, and enable safe release strategies for model versions.

Key responsibilities include orchestrating training and inference workloads on Databricks, embedding security into