Description:

The SRE/HPC Engineer at Fluidstack is responsible for ensuring peak performance of the GPU infrastructure and providing top-tier support to customers.
Responsibilities include deploying new clusters on a monthly basis, automating processes to support scalability, and offering client-facing support for tasks like GPU debugging and performance optimization.
The role involves working with top AI companies like Poolside, Meta, Modal, and Reka.

Requirements:

Experience in HPC systems, System Administration, SRE, or DevOps.
Proficiency in managing large-scale workloads with orchestrators like Slurm or Kubernetes.
Ability to automate processes for bare-metal machines and containers using tools like Ansible, Bash, or Python.
Familiarity with shared storage platforms such as NFS, DDN, Vast, CephFS, etc.
Experience in provisioning large-scale clusters and networks with tools like BCM, UFM.
Knowledge of large-scale GPU systems, including working with Nvidia GPUs and Infiniband networks.
Must be a fast learner, adaptable, and passionate about Fluidstack’s mission.

Opportunity to work with top AI companies in the industry.
Chance to contribute to the growth and scalability of Fluidstack's GPU infrastructure.
Client-facing role providing exposure to diverse challenges and problem-solving opportunities.
Continuous learning and development in a dynamic and innovative environment.