Principal Infrastructure Developer

Electronic Arts is seeking a Principal Infrastructure Developer to join their team. As a Principal Infrastructure Developer, you will be responsible for the entire graphics processing unit (GPU) infrastructure that supports the company's research efforts.

You will define the technical direction for GPU operations and infrastructure architecture. Additionally, you will be responsible for setting up infrastructure as code, managing permissions, and resolving infrastructure-related issues.

Responsibilities:

• Manage GPU operations across AWS
• Develop planning layers from scratch
• Diagnose node and GPU failures quickly and thoroughly
• Provide support to researchers through dedicated support windows and channels
• Instrument units to measure utilisation, queue lengths, task success rates, and costs
• Collaborate with external partners and internal teams on shared services and incident management
• Develop operational and decision-making documentation for new employees

Requirements:

• At least 8 years of experience in production infrastructure operations
• In-depth knowledge of AWS, including EC2, EKS, IAM, security, VPCs, networks, S3, and FSx
• Experience planning, managing, and diagnosing GPU clusters in AWS
• Expertise in scripting and automation using Python, PowerShell, bash, or equivalent
• Knowledge of infrastructure as code (Terraform or equivalent)
• Familiarity with GPU scheduling layers (Slurm, Kubernetes, Ray, etc.)
• Observability practices with Grafana, Prometheus, or equivalent

This role is hybrid, requiring three days per week in the office at Montreal, Redwood City, or Vancouver.

Salary:

• British Columbia: $169,500 - $242,600 CAD
• California: $193,100 - $296,500 USD

Benefits:

• Comprehensive benefits packages, including paid time off, sick time, medical/dental/vision insurance, life insurance, disability insurance, and retirement plans