Devops, Distributed GPU Cloud Engineer

19 مايو، 2023

0 0 2 دقائق

الوصف الوظيفي

What You’ll Do

Design, implement, and manage deployment of highly available services
Lead the implementation of a CI/CD pipeline for existing repositories
Contribute to design, implementation, testing, and documentation of orchestration and other services supporting GPU Cloud
Contribute to major cloud features that enable our customers to easily run large-scale AI research projects
Develop and maintain tools for automation, deployment, monitoring, and operations of our cloud infrastructure
Develop and maintain the infrastructure as code, including the design and implementation of automated infrastructure deployment and configuration management
Provide guidance and support to development teams in the implementation of DevOps best practices
Participate in building orchestration software within our data centers for launching/terminating VMs, networking, and attaching storage
Ensure correctness of software through high quality automated testing and setting standards for other developers
Stay current with emerging trends and technologies related to DevOps, cloud computing, and infrastructure automation
Linux distributions and package managers
Linux user management
VM management
GitLab administration
Network building, configuration and maintenance
Monitoring tool setup and maintenance (i.e.:Grafana)
Automatic backup setup and maintenance

Strong engineering background – EECS preferred, Mathematics, Software Engineering, Physics
5+ years of experience in DevOps or related field
Experience with the implementation of CI/CD pipelines for existing repositories using tools like Github Actions or Buildkite
Strong experience with infrastructure as code (IaC) and related tools such as Ansible, CloudFormation or Terraform
Proficient in one or more programming languages such as Python, Node.JS or Go
Familiarity with cloud computing platforms such as AWS, GCP, or experience deploying a web application on a a private cloud platform
Experience with containerization technologies such as Docker or Kubernetes

Nice to Have

Experience in a leadership role, with the ability to lead and mentor junior team members
Have experience developing major features from conception to production using a Python web application
Experience integrating with storage systems, data center networking, virtualization
Experience in the machine learning or computer hardware industry
Cloud platform management
Shell scripting
Container orchestration tool familiarity
Knowledge of cluster schedulers
Designing cloud deployment strategy for Riva and NeMo LLM and any other LLM model services: helm charts, k8s operators, etc
slurm, zabbix, rockclusters, proxmox for GPU orchestration and distributed full power in one computing solutions
Experience with GPU computing systems
Experience with Nvidia (Cuda) drivers.
Tensor flow, PyTorch experience is advantageous