Senior MLOps Engineer(GCP)

Job Description: Senior MLOps Engineer

Location: Remote from Spain (Spanish employment contract)

About the Role

We are looking for an engineer to build and maintain ML/AI infrastructure on Google Cloud (GCP).

The main focus: make sure models (including LLMs) run reliably, securely, and in production — not just in researchers’ notebooks.

This is not a Data Scientist and not a pure DevOps. It’s a role at the intersection of Cloud + Data + ML.

Key Responsibilities

• Build and manage ML pipelines from training to production deployment (Vertex AI, Dataproc, Airflow/Cloud Composer).
• Set up CI/CD for ML: model deployment, testing, and updates automation.
• Manage infrastructure as code (Terraform, GitLab CI/CD).
• Containerization & orchestration (Docker, Kubernetes/GKE).
• Monitor models in production (Prometheus, Grafana, MLflow).
• Integrate LLMs (HuggingFace, OpenAI API, Gemini, LLaMA).
• Handle data workflows: BigQuery, Spark/Databricks, ETL pipelines.
• Ensure security & compliance (IAM, access control, audits).

Tech Stack

• Cloud: Google Cloud (Vertex AI, BigQuery, Dataproc, Composer, GKE).
• ML: TensorFlow, PyTorch, HuggingFace, MLflow.
• Data: Spark, Databricks, Airflow.
• Infra: Terraform, Docker, Kubernetes.
• CI/CD: GitLab, Jenkins (or similar).
• Monitoring: Prometheus, Grafana.
• Security: IAM, cloud/ML best practices.

What You Should Have

• Hands-on production experience with GCP (not just courses or theory).
• Solid understanding of how ML works in production: logging, monitoring, model versioning & rollback.
• Strong experience with Kubernetes & Terraform (must-have).
• Solid Python skills (ideally with ML libraries).
• Experience with LLMs is a strong plus.

Why This Role Is Challenging

• Many companies stop at PoC — here you’ll need to build real production-grade pipelines.
• LLM integration is still raw — you need a cool head to deploy & monitor them properly.
• GCP is less common than AWS/Azure — experience with Google Cloud is highly valued.

In one sentence:

This role is for an engineer who can turn researchers’ ML code into stable production services on GCP.