Description
Roku is hiring a Senior Software Engineer, MLOps/DevOps for its Advertising Performance team to design and operate scalable cloud infrastructure for machine learning workloads across AWS and GCP. The role covers GPU/TPU training and inference environments, CI/CD for ML models, low-latency inference infrastructure, observability, incident response, and automation across the ML lifecycle. It requires extensive DevOps, SRE, and ML infrastructure experience, strong programming and cloud-platform skills, and familiarity with Kubernetes, Spark, Airflow, Terraform, and related tools. The role follows a hybrid schedule with office attendance Monday through Thursday and flexible Fridays.
