MLOps Engineer: Train, Serve & Observe ML Systems

Electronic Arts (EA); Redwood City (CA); Hybrid; USD 165,000 - 256,000; Full time

Benefits offered by this job

PTO 3 weeks/year

Sick time 80 hours/year

16 paid holidays/year

Parental leave 10 weeks

Health/dental/vision insurance

Life and disability insurance

401(k) plan

Bonus potential

Job summary

Electronic Arts seeks an experienced MLOps Engineer to collaborate with researchers, optimize model training, and shape data needs for training. You will work with the Lead Infrastructure Engineer to translate requirements into durable, scalable processes and ensure the fleet runs reliably.

You will report to the Head of Data & Infrastructure and contribute to a hybrid work model across locations in Redwood City, Montreal, or Vancouver, with a focus on safety, reproducibility, and performance.

Qualifications

• 8+ years of engineering experience.
• 4+ years building and operating ML systems in production.
• Expertise in Python.
• Experience with PyTorch and inference servers.
• Experience with distributed data processing for ML (Ray, Spark).
• Observability skills with Prometheus/Grafana.
• AWS for ML workloads (EC2, S3, EKS).
• Containerization and CI/CD for ML artifacts.

Responsibilities

• Optimize how our models train and serve.
• Build the streaming data path from lake to GPU.
• Make our data formats a deliberate choice.
• Instrument GPU utilization, queue depth, and job success rate; monitor cost per experiment and serving latency percentiles.
• Land training and experiment telemetry into a single trusted repository.

Skills

Python

PyTorch

Distributed data processing

Observability

AWS for ML workloads

Tools

Ray

Spark

Prometheus

Grafana

EC2

S3

EKS

Job description

Electronic Arts seeks an experienced MLOps Engineer to collaborate with researchers, optimize model training, and shape data needs for training. You will work with the Lead Infrastructure Engineer to translate requirements into durable, scalable processes and ensure the fleet runs reliably.

You will report to the Head of Data & Infrastructure and contribute to a hybrid work model across locations in Redwood City, Montreal, or Vancouver, with a focus on safety, reproducibility, and performance.