ML Platform Engineer (all)

Let me introduce... With Sonia, doctors are successful doctors. We create and deploy AI enhanced solutions that make doctors’ lives easier, patients’ care better, and healthcare systems more efficient.

Support and enhance our Kubernetes-based infrastructure in cloud environments, running both ML/LLM workloads and general applications
Deploy and optimize LLM inference systems
Design, build, and improve MLOps/DevOps pipelines to support the entire development lifecycle
Manage GPU scheduling and autoscaling with Kubernetes-native tooling
Ensure observability and alerting across the platform
Operate and troubleshoot supporting infrastructure
Contribute to platform reliability, security, and performance through automation and best practices

5+ years of experience in MLOps or SRE
Strong hands-on Kubernetes experience, including GitOps (Flux or ArgoCD), Kustomize, Helm and production troubleshooting
Familiarity with LLM inference deployment and optimization in Kubernetes (e.g., vLLM, LMCache, llm-d)
Experience with MLOps supporting tools such as MLflow or Argo Workflows
Understanding of GPU resource orchestration in Kubernetes environments
Profound knowledge of observability tools, such as VictoriaMetrics, VictoriaLogs and Grafana
Knowledge of database and broker administration (PostgreSQL, Redis and RabbitMQ)
Solid scripting skills in Python
Comfortable working with cloud platforms (OVHcloud, AWS, GCP or Azure)
Experience with audio ML models or real-time inference
Exposure to CI/CD practices tailored for ML systems
Familiarity with Kubernetes networking, security, or performance tuning

Full ownership of a mission-critical platform
A team that values curiosity, learning, and experimentation
Remote-first setup with the option to work in our Berlin office
Competitive salary depending on experience
Work on AI infrastructure that directly impacts healthcare innovation