Location
Amsterdam, Berlin, London, Prague, or remote within Europe.
About the company
A global AI cloud provider building a full-stack platform for training and deploying models at scale, with large GPU fleets and R&D hubs across Europe, the UK, North America and Israel.
The team focuses on an inference platform serving text, vision, audio and multimodal foundation models reliably under heavy load.
Responsibilities
- Own reliability, performance and observability across the inference stack, including metrics, logs and tracing pipelines.
- Tune Kubernetes autoscaling and GPU efficiency, and build resilient infrastructure with Terraform.
- Harden request routing and retry logic, and automate incident detection, isolation and remediation.
- Lead post-mortems and collaborate with software engineers to make reliability a built-in feature.
Requirements
- Strong hands-on experience with Kubernetes, Prometheus, Grafana, Terraform and infrastructure-as-code.
- Scripting proficiency in Python or Bash, plus solid knowledge of alerting and SLOs for high-throughput APIs.
- Production experience with distributed backends and debugging performance from kernel to application layer.
- Experience with GPU-heavy workloads such as vLLM, Triton or Ray, and a background in MLOps or model-hosting platforms.