Zyphra
Platform Engineer
San Francisco · On-site · Full-time · Posted Mar 17, 2026
- Visa sponsorship
- Relocation and immigration support on a case-by-case basis.
About the job
Designs and maintains reliable, observable, secure, and scalable infrastructure for ML workloads and research and production deployments.
What you'll do
- Build observability systems and manage infrastructure and CI/CD in partnership with engineering teams.
- Design resilient build, deployment, and secure release systems with auditability and rollback support.
- Lead incident response, root-cause analysis, and postmortems; improve system reliability and performance.
What they ask for
- Experience in high-performance compute environments and hyperscaler cloud environments.
- Experience with infrastructure as code, run-books, disaster recovery planning, change management, and fault tolerance.
- Familiarity with containers and their integration with scheduling systems.
- Experience with deployment strategies at scale, reproducible experimental workloads, compliance and audit standards, and load testing or chaos engineering.
Nice to have
- Familiarity with software release engineering for ML/AI systems.
- Exposure to backend development for ML model serving, such as vLLM, Ray, SGLang, or Triton.
Tools
AWS, GCP, Terraform, Ansible, Docker, Apptainer, Kubernetes, Slurm
AI focus
Supports ML/AI infrastructure, including GPU management, workload optimization, reproducible ML workloads, and potentially backend development for ML model serving.
Model-written summary · the original posting is authoritative