MLOps Engineer

Xebia is a global AI-first, digital transformation, and engineering partner. With over 25 years of experience and a team of 5,000 professionals across 16 countries, we help organizations design and build scalable products, platforms, and data-driven solutions.

You will be:

• owning the machine learning platform infrastructure as part of the Data Infrastructure team, which operates the company-wide GCP data and ML platform,
• owning Vertex AI, Kubeflow Pipelines, and GKE-based training and serving infrastructure, keeping it reliable, secure, and cost-efficient,
• building templates, CI/CD, and self-service workflows that enable data scientists to ship models without filing tickets,
• managing the model lifecycle, including model registry, versioning, deployment, rollback, and monitoring for drift and serving health,
• supporting feature and training data flows across BigQuery and Airflow, in collaboration with data platform engineers,
• managing infrastructure through GitOps with Config Sync, applying least-privilege access for PII and policy-tagged data,
• managing GPU and compute usage, quotas, and per-team spend visibility,
• taking end-to-end ownership of the platform that vertical teams use to build, train, and serve their models.

Your profile:

• 5+ years in MLOps, ML platform or platform engineering, including at least 2 years owning production ML infrastructure end to end,
• hands-on production experience with GCP, specifically Vertex AI, GKE and BigQuery,
• strong Kubernetes expertise, including running and debugging clusters independently,
• pipeline orchestration with Kubeflow Pipelines or Vertex AI Pipelines, plus Airflow,
• GitOps for infrastructure (Config Sync or similar) and CI/CD for model deployment,
• production-grade Python skills, including testing and packaging,
• a platform mindset: experience building documented, self-service tooling for other teams, with SLOs,
• ability to work independently as the single owner of a company-wide platform.

Work from the European Union region and a work permit are required.

Nice to have:

• streaming features or online inference using Dataflow/Flink and Kafka,
• feature stores such as Vertex AI Feature Store or Feast, and model monitoring tooling,
• LLM serving and evaluation on Vertex AI or similar platforms,
• GPU scheduling and cost optimisation on GKE,
• Data governance experience (DataHub, policy tags) and privacy regulations such as GDPR or DPDP,
• experience with high-scale consumer products.