Platform Engineer

As a Platform Engineer, you’ll be responsible for designing and maintaining the systems that keep Zyphra’s infrastructure robust, observable, secure, and scalable. Your work will be essential to ensuring the reliability and reproducibility of ML workloads, the safety and control of deployments, and the long-term maintainability of our compute environments.

• Building and improving observability systems (monitoring, logging, alerting)

• Managing Infrastructure as a Service across the stack along with CI/CD in close partnership with engineering teams

• Designing resilient build and deployment systems across research and production environments

• Implementing secure release processes with strong auditability and rollback support

• Collaborating closely with ML engineers, DevOps, and infra teams to improve system reliability and performance

• Leading incident response, root-cause analysis, and postmortems with a focus on learning and prevention

• This role is ideal for someone who loves building systems that make other teams faster, safer, and more productive

• Experience in high-performance compute environments, such as ML clusters or GPU farms as well as hyperscaler cloud environments (i.e. AWS, GCP, etc.)

• Background in infrastructure as code (i.e., Terraform, Ansible, etc.)

• Familiarity with containers (i.e., Docker, Apptainer) and their integration with scheduling systems (i.e., Kubernetes, Slurm)

• Familiarity with software release engineering for ML/AI systems is a plus

• Experience managing run-books, DRP, change management, and general fault tolerance

• Experience with deployment strategies at scale

• Experience designing reliable environments for experimental workloads and reproducible runs

• Knowledge of compliance and audit standards in deployment and system security

• Experience with load testing, fault injection, and chaos engineering to harden systems under stress

• Passion for building tooling that makes infrastructure invisible and reliable for end users

• Experience with infrastructure as code (e.g., Ansible, Terraform)

• Prior work supporting ML/AI infrastructure, including GPU management and workload optimization

• Exposure to backend development for ML model serving (i.e., vLLM, Ray, SGLang, Triton)