ML Deployment Engineer

XpertDirect; München; Vor Ort; EUR 90.000 - 130.000; Vollzeit

Zusammenfassung

XpertDirect seeks an ML Deployment Engineer in Munich to bridge ML experimentation and production. You will design and operate deployment pipelines, ensure reliable model serving, and build scalable inference services that run on Kubernetes across AWS/GCP environments.

You will implement canaries, blue-green strategies, and robust observability with MLflow, KServe, and Docker-based tooling, collaborating with ML engineers to productionise models with minimal infrastructure management.

Qualifikationen

• 3+ years in MLOps, ML Engineering, ML Infrastructure, Platform Engineering, or similar roles.
• KServe or comparable model-serving technology.
• AWS and/or GCP.
• Strong understanding of production ML systems.

Aufgaben

• Build production deployment pipelines for machine-learning models.
• Deploy and operate model-serving workloads on Kubernetes.
• Build scalable inference services using KServe.
• Containerise ML workloads using Docker.
• Develop deployment tooling and automation in Python.
• Manage model versions, artefacts, and deployment workflows with MLflow.
• Build CI/CD pipelines for testing and releasing ML services.
• Deploy workloads across AWS and/or GCP environments.
• Implement rollout, rollback, and model versioning strategies.
• Improve deployment reliability, scalability, and observability.
• Automate the path from approved model to production endpoint.
• Collaborate with ML Engineers to productionise new models.

Kenntnisse

MLOps

ML Engineering

KServe

AWS/GCP

Python

Docker

Canary deployments

Tools

Docker

Kubernetes

MLflow

Python

Terraform

Jobbeschreibung

Deep Tech | MLOps | Model Deployment | Model Serving | ML Infrastructure

Our client, a growing Deep Tech / AI company based in Munich, is looking for an ML Deployment Engineer to build the deployment layer that takes machine-learning models from experimentation into scalable, reliable production services.

You'll work at the intersection of ML Engineering, MLOps, and Platform Engineering, creating the tooling and infrastructure that makes model deployment repeatable, observable, and production-ready.

What You'll Work On

• Build production deployment pipelines for machine-learning models
• Deploy and operate model-serving workloads on Kubernetes
• Build scalable inference services using KServe
• Containerise ML workloads using Docker
• Develop deployment tooling and automation in Python
• Manage model versions, artefacts, and deployment workflows with MLflow
• Build CI/CD pipelines for testing and releasing ML services
• Deploy workloads across AWS and/or GCP environments
• Implement rollout, rollback, and model versioning strategies
• Improve deployment reliability, scalability, and observability
• Automate the path from approved model to production endpoint
• Collaborate with ML Engineers to productionise new models without requiring them to manage the underlying infrastructure

Core Skills

• 3+ years in MLOps, ML Engineering, ML Infrastructure, Platform Engineering, or similar roles
• KServe or comparable model-serving technology
• AWS and/or GCP
• Strong understanding of production ML systems

Nice to Have

• NVIDIA Triton Inference Server
• Ray Serve
• PyTorch / TensorFlow
• Terraform
• Canary or blue-green deployments
• GPU-enabled inference workloads
• Model monitoring and drift detection
• Experience operating real-time inference APIs