SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
DeepL is seeking a Staff Platform Engineer to join its Hybrid Platform Engineering track, which builds and operates the compute infrastructure underpinning DeepL's AI products and research. You will take ownership of the reliability, capacity, and cost efficiency of DeepL's hybrid compute infrastructure spanning both on-premises GPU clusters (powering research and production inference) and AWS regions globally.
Key responsibilities include:
- Owning the reliability and cost efficiency of hybrid compute infrastructure (CPU and GPU) across AWS and on-prem, including hardware lifecycle management
- Shaping platform architecture by bringing strong technical positions and committing to decisions that serve the best outcomes
- Designing, building, and operating production-grade Kubernetes clusters across cloud and on-prem infrastructure from design through long-term operation
- Driving technical work to deepen the hybrid model, unifying how workloads run across on-prem and cloud environments
- Defining infrastructure-as-code standards and strengthening observability and security across the platform
- Mentoring engineers across the infrastructure track and raising the technical bar through design reviews and architecture decisions
- Building consensus across engineering teams and shaping platform direction at the track level
- Leading incident response for hybrid infrastructure and sustaining an on-call rotation
Required qualifications:
- Deep, hands-on Kubernetes expertise with experience designing, building, and operating clusters at scale through their full lifecycle
- Depth at scale in either public cloud or on-prem infrastructure, with AWS experience preferred and bare-metal/data centre experience strongly valued
- Networking and Linux expertise sufficient to debug issues from container through host to network edge
- Infrastructure-as-code experience with Terraform or equivalent, and GitOps delivery with tools like ArgoCD in production environments
- Software engineering skills in Go or Python at the level of building tooling other engineers rely on
- Track record of technical ownership spanning multiple teams, with ability to create clarity across team boundaries and carry other engineers with you
- Strong incident-management and reliability-engineering discipline with an ownership mindset
Bonus qualifications include GPU infrastructure experience at scale, distributed storage systems (e.g., Ceph), experience serving researchers/data scientists, and BGP-level networking in hybrid or on-prem environments.
DeepL is a global AI company founded in 2017 with ~1,000 employees across 228 markets, backed by Benchmark, IVP, and Index Ventures. The company is focused on building secure, intelligent AI solutions with a strong culture emphasizing innovation, growth, and well-being.
