MLOps Engineer

Role: MLOps Engineer

Job Ref Id: JD-1283

Competence level: 5+ years

Start date: 2026-12-01

End date: 2027-06-01

Extension option: Yes

Scope: 100%

Location: Stockholm

Remote work: Remote

Language: English

Application deadline: 2026-11-01

Number of consultants: 1

Security clearance: No

About the assignment

We are looking for a MLOps contractor to own our ML platform infrastructure. You will sit in the Data Infrastructure team, which runs the GCP data and ML platform for the whole company. Vertical teams own their models and use-case logic; you own the platform they build on: training, serving, pipelines, and the tooling around them. This is one role covering company-wide ML infra, so you'll be the go-to person from day one.

Responsibilities

• ML platform on GCP: Vertex AI, Kubeflow Pipelines and GKE-based training and serving infrastructure, kept reliable, secure and cost-efficient.
• Paved roads for ML teams: templates, CI/CD and self-serve tooling so data scientists ship models without filing tickets.
• Model lifecycle: registry, versioning, deployment, rollback, and monitoring for drift and serving health.
• Data integration: feature and training data flows across BigQuery, Airflow together with the data platform engineers.
• Infrastructure as code: everything managed through GitOps with Config Sync, with access following least privilege for PII and policy-tagged data.
• Cost and capacity: GPU/compute usage, quotas, and spend visibility per team.

Requirements

Mandatory

• 5+ years in MLOps, ML platform or platform engineering, with at least 2 years owning production ML infrastructure end to end.
• Hands-on GCP: Vertex AI, GKE and BigQuery in production, not only as skill tags.
• Kubernetes depth: you've run and debugged clusters yourself, not just deployed to them.
• Pipeline orchestration with Kubeflow Pipelines or Vertex AI Pipelines, plus Airflow.
• GitOps for infrastructure (Config Sync or similar) and CI/CD for model deployment.
• Python to production standard, including tests and packaging.
• Proven platform thinking: you've built tooling other teams self-serve on, with docs and SLOs.
• Comfortable working independently as the single owner of a company-wide platform.

Nice-to-have

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

Originally posted on Himalayas