Expert for Kubernetes | DevOps | NVIDIA GPUs | Kong | max. 90€ all in

Project overview

For our client, we are currently looking for an

Expert for Kubernetes | DevOps | NVIDIA GPUs | Kong | max. 90€ all in - Düsseldorf, Kaarst (20%) | Remote (80%)

Framework Information

Your Qualifications

• Reference: 180526
• Start: 2026-11-02
• Duration: until 12/31/2026 + annual extension
• Location: Düsseldorf, Kaarst (20%) | Remote (80%)
• Scope of work: full-time
• Languages: German
• Experience with Kubernetes – focus of the activity, especially in the setup, operation, and administration of Kubernetes clusters
• Experience with GPU infrastructures, ideally with NVIDIA GPUs, including CUDA, cuDNN, and multi-GPU scheduling
• Knowledge of working with API managers, e.g., Kong
• Experience with Model Serving, e.g., with NVIDIA Triton Inference Server
• Knowledge in the area of Service Mesh, ideally with Istio
• Experience with API security, especially rate limiting and audit logging
• Knowledge in the area of Model Monitoring and Observability, especially with Grafana for creating dashboards
• Experience with workflow orchestration in the ML environment, e.g., with MLflow
• Knowledge in the operation and administration of Keycloak and cert-manager

Responsibilities

Interested?

We look forward to receiving your comprehensive application documents in electronic form.

Your Contact Persons

Mr. Jesse Lüder | #gerneperdu

P: [contact removed]

E: [contact removed]

• Setup and optimization of a scalable NVIDIA GPU infrastructure including CUDA and CuDNN configuration as well as multi-GPU scheduling for parallel AI workloads.
• Implementation and management of an API manager (e.g., Kong) for central control and security of microservice interfaces.
• Setup of a Model Serving framework (e.g., Triton Inference Server) for high-performance deployment and versioning of AI models.
• Containerization and orchestration of AI services with Kubernetes, including deployment strategies, scaling, and cluster monitoring.
• Integration of a Service Mesh (e.g., Istio) for implementing traffic management, service discovery, and zero-trust communication.
• Implementation of API security mechanisms such as rate limiting, authentication, authorization, and audit logging.
• Setup of a model monitoring system with Grafana and Prometheus for monitoring model performance and system metrics.
• Introduction of a workflow orchestration solution (e.g., MLFlow) for managing training, validation, and deployment pipelines.
• Administration of operational components such as Keycloak (Identity & Access Management) and cert-manager (TLS certificate management) to ensure productive operation.

Project details

• Start: 2026-11-02
• Duration: bis 31.12.2026 + Verlängerung Jahresweise
• Remote: hybrid
• Location: Düsseldorf, DE
• Languages: German