Sr. Software Engineer – MLOps

About Roku

Roku pioneered streaming to the TV. We connect users to the streaming content they love, enable content publishers to build and monetize large audiences, and provide advertisers with unique capabilities to engage consumers. Roku streaming players and Roku TV™ models are available around the world through direct retail sales and licensing arrangements with TV brands and pay TV operators. With tens of million players sold across many countries, thousands of streaming channels and billions of hours watched over the platform, building scalable, highly available, fault-tolerant, big data platform is critical for our success. 

About the role

You will join the Roku Pay Data Engineering team and will work to accelerate our efforts around anomaly detection and recommendation systems within the payments ecosystem. In this pivotal role, you will work closely with Data Scientists to spearhead the development of robust, scalable, and efficient infrastructure essential for model serving, training, feature engineering, and data processing. Your expertise will be crucial in building and maintaining ML models to serve real-time alerting and insights. 

What would you be doing

• Design and implement automation for model training, testing, validation, and deployment, integrating CI/CD practices to ensure smooth and consistent delivery of ML models 

• Deploy machine learning models into production environments, establishing monitoring and alerting systems to track model performance, health, and data drift 

• Optimize compute resources for cost and performance efficiency. 

• Ensure high availability and scalability of ML infrastructure. 

• Collaborate with cross-functional teams (Data Platform, Data Science) to streamline end to end ML production flow efforts 

What you have done and what you bring

• 5+ years of experience in the design of large scale & production ML Infra systems or Distributed Systems. 

• Experience in deploying and maintaining full lifecycle of ML models 

• Experience with orchestration tools like Airflow, KubeFlow, MLFlow, Vertex AI etc. 

• Proficiency in Cloud computing (e.g. GCP, AWS) and containerization technologies (e.g. Kubernetes, Docker) 

• Strong Python skills and familiarity with following open source ML libraries: Scikit-learn, Pandas, PyTorch, Tensorflow, and deep learning packages (e.g. RNN) 

• Familiarity with ML optimization models with standard frameworks/solvers (k-means, decision trees, ensemble models) 

• Familiarity with recommendation systems and other ML techniques 

• Familiarity with big data technologies and streaming architecture, data pipelines, Spark, SQL, etc. 

• Experience in payments domain is preferred 

• BS (or higher) degree in Computer Science, or a related field. 
  

#LI-AR3