Machine Learning Systems Administrator / DevOps Engineer: Training Clusters & Inference Cloud
Core
Maintain and build core infrastructure for large-scale model training clusters and production inference cloud services on heterogeneous compute.
Role type
Senior IC ML Systems Administrator / DevOps Engineer
Builds
Reliable, secure, and fast training clusters and inference services for model deployment
Domain
AI/ML Infrastructure, High-Performance Computing, Cloud Systems
Deliverable
infrastructure
Required skills
Linux system administration, Bash/Python scripting, Container orchestration (Kubernetes, Slurm), Monitoring & alerting (Prometheus, Grafana), Incident response, Debugging (compute/storage/networking), Infrastructure as Code (Terraform, Ansible)
Preferred skills
GPU management, LLM serving stacks (vLLM, SGLang, Ray), High-performance networking (InfiniBand, RoCE), Distributed storage, Cloud platforms (AWS, Azure, GCP), Apptainer integration
Technologies
Linux, Docker, Kubernetes, Slurm, Prometheus, Grafana, Terraform, Ansible, vLLM, SGLang, Ray, NVIDIA CUDA, AMD ROCm, TPUs, InfiniBand, RoCE, AWS, Azure, GCP, Apptainer
Responsibilities
Administer and automate Linux-based cluster environments; Deploy and scale inference services across heterogeneous accelerators; Manage container orchestration and CI/CD pipelines; Build observability systems for production services; Plan capacity and track utilization/cost; Secure production systems through secrets management and network policies
Seniority
Senior, hands-on IC
