Staff Platform Engineer

Overview

As a Hybrid Platform Engineer, you own the reliability and cost efficiency of DeepL’s compute infrastructure across AWS and on‑prem, including GPU clusters for research. You will shape and operate the platform that powers both product and research teams, unifying workloads across environments. This role combines hands‑on engineering with architectural decisions to scale our hybrid infrastructure. You’ll mentor peers, lead incident response, and drive a secure, observable foundation for the organisation.

Pay / Benefits

• hybrid work with flexible hours
• virtual shares ownership
• 30 days annual leave
• regular in‑person and team events
• monthly Hack Fridays
• international team with diverse backgrounds

Responsibilities

• Own reliability, capacity, and cost efficiency of hybrid compute infrastructure (AWS and on‑prem)
• Shape platform architecture with peer input and clear positions
• Design, build and run production-grade Kubernetes clusters across cloud and on‑prem
• Advance the hybrid model to unify workloads between on‑prem hardware and cloud (especially AWS)
• Define infrastructure‑as‑code and platform standards, enhance observability and security
• Mentor engineers across the infrastructure track and maintain high technical standards
• Build consensus across teams and translate it into shipped platform changes
• Lead incident response for hybrid infrastructure and sustain an on‑call rotation

Key requirements

• Deep, hands‑on Kubernetes expertise (design, build, operate at scale)
• Experience with either public cloud or on‑prem at scale; AWS preferred; bare‑metal/data centre valued
• Strong networking and Linux debugging across container/host/network edge
• Infrastructure as code (Terraform or equivalent) and GitOps with ArgoCD
• Software engineering skills in Go or Python; building practical tools
• Track record of technical ownership spanning several teams
• Strong incident management and reliability engineering discipline
• Bonus: GPU infra for training/inference, distributed storage (Ceph), platform work for researchers/data scientists, and advanced networking (BGP) in hybrid/on‑prem

• collaboration
• leadership and mentoring
• problem‑solving and decision making
• Kubernetes at scale
• AWS and on‑prem infrastructure
• Networking and Linux depth

…