MLOps Engineer — Real-Time AI Platform Lead

Aristocrat; Cerdanyola del Vallés; Hybrid; EUR 60,000 - 90,000; Full time

Job summary

Aristocrat is seeking an experienced MLOps Engineer to build and own the platform that brings ML and AI models from prototype to production for Product Madness. You will deploy real-time services, manage CI/CD and infrastructure, and enable ML engineers to ship and monitor models with confidence.

You will collaborate with ML engineers and data scientists to smooth the path from experimentation to production while ensuring reliability and observability across environments.

Qualifications

• 4+ years in ML infrastructure, ML platform, or MLOps roles, ideally within a dedicated team.
• Strong software and systems foundations with a CS/engineering background.
• Experience serving models as real-time low-latency services and handling caching/concurrency.
• Hands-on with Docker, Kubernetes, CI/CD and cloud platforms (GCP/AWS/Azure).

Responsibilities

• Deploy models as reliable real-time APIs with focus on latency and throughput.
• Design and own serving/deployment infra: containers, orchestration, and serving frameworks.
• Build and maintain CI/CD pipelines for models and retraining workflows.
• Own experiment tracking and model registry with versioning and governance.
• Develop monitoring and observability for ML systems (metrics, logs, traces).
• Harden systems for reliability (retries, timeouts, rollback) and ensure reproducibility.

Skills

Docker

Kubernetes

CI/CD

Python

Airflow

MLflow

Weights & Biases

Tools

KServe

Triton

BentoML

Prometheus

Grafana

Weights & Biases (W&B)

Job description

Aristocrat is seeking an experienced MLOps Engineer to build and own the platform that brings ML and AI models from prototype to production for Product Madness. You will deploy real-time services, manage CI/CD and infrastructure, and enable ML engineers to ship and monitor models with confidence.

You will collaborate with ML engineers and data scientists to smooth the path from experimentation to production while ensuring reliability and observability across environments.