MLOps Engineer — Hybrid ML Training & Serving at Scale

Electronic Arts (EA); Montreal (administrative region); Hybrid; CAD 141,000 - 204,000; Full time

Job summary

Electronic Arts is hiring an MLOps Engineer to collaborate with researchers and optimize model training and deployment. The role involves shaping data needs, building robust pipelines, and partnering with infrastructure leadership to ensure reliable, scalable ML workloads across a hybrid onsite setup.

The position reports to the Head of Data & Infrastructure. You will work on streaming data paths, deliberate data format choices, and telemetry instrumentation to measure GPU utilization, latency,

Qualifications

• 8+ years of engineering experience.
• 4+ years building and operating ML systems in production.
• Expertise in Python.
• Comfort reading and modifying framework-level code — PyTorch and an inference server such as vLLM, SGLang or TensorRT-LLM.
• Demonstrated model or inference optimization work.
• Hands-on experience with distributed data processing for ML (including Ray, Spark, or equivalent) and columnar and ML-native formats (such as Parquet, Arrow, Lance, or similar).
• Practical observability skills including Prometheus, Grafana, or equivalent.
• Working knowledge of AWS compute and storage as they apply to ML workloads — EC2 GPU instances, S3 and its performance tiers, EKS.
• Experience with containerization and CI/CD for ML artifacts and a reproducibility instinct.

Responsibilities

• You will optimize how our models train and serve.
• You will build the streaming data path from lake to GPU.
• You will make our data formats a deliberate choice.
• You will instrument all of it including Per-job GPU utilization, queue depth and job success rate, cost per experiment, data-loader stall time, and serving latency percentiles.
• You will land the training and experiment telemetry into one place researchers and planners trust.

Skills

Python

Model training

Distributed ML systems

Observability

AWS

Containerization

Education

Bachelor's degree in CS/CE or related field

Tools

PyTorch

vLLM

TensorRT-LLM

Ray

Spark

Parquet/Arrow

Prometheus

Grafana

EC2

S3

EKS

Docker

CI/CD

Job description

Electronic Arts is hiring an MLOps Engineer to collaborate with researchers and optimize model training and deployment. The role involves shaping data needs, building robust pipelines, and partnering with infrastructure leadership to ensure reliable, scalable ML workloads across a hybrid onsite setup.

The position reports to the Head of Data & Infrastructure. You will work on streaming data paths, deliberate data format choices, and telemetry instrumentation to measure GPU utilization, latency,