Staff Platform Engineer
Core
Design, build, and operate the hybrid compute layer (Kubernetes control plane, networking, storage) powering DeepL's AI products and research across AWS and on-prem GPU clusters.
Role type
Staff Platform Engineer (Infrastructure)
Builds
Hybrid Kubernetes clusters, on-prem GPU infrastructure, and core cluster services for production inference and research workloads.
Domain
Cloud Infrastructure / Hybrid Cloud / AI Research Infrastructure
Deliverable
infrastructure
Required skills
Kubernetes cluster design and operation at scale, Linux debugging, Networking (including BGP), Infrastructure as Code (Terraform), GitOps (ArgoCD), Software engineering (Go or Python), Incident management, Reliability engineering
Preferred skills
GPU infrastructure for training/inference, Distributed storage systems (Ceph), Serving researchers/data scientists, Networking at BGP level in hybrid environments
Technologies
Kubernetes, AWS, Terraform, ArgoCD, Go, Python, Linux, BGP, Ceph
Responsibilities
Take ownership of reliability, capacity, and cost efficiency of hybrid compute infrastructure; Shape platform architecture and make technical decisions; Design and operate production-grade Kubernetes clusters; Drive the hybrid model unification between on-prem and cloud; Define infrastructure-as-code standards and strengthen observability/security; Mentor engineers and raise the technical bar; Lead incident response and drive remediation.
Seniority
Staff, hands-on IC with mentorship and architectural leadership
