Senior Machine Learning Engineer - Advertising Engineering

What does the team work on?

A reliable, privacy-safe, and scalable data foundation is critical to Roku's rapidly growing advertising business. The Ad Data Activation organization builds the identity systems, device graph pipelines, audience platforms, insights tooling, and data collaboration capabilities that power targeting, measurement, and reporting across Roku Ads.

What is the role?

Roku is hiring a Senior Machine Learning Engineer within the Ad Data Activation organisation to drive intelligence across advertising data systems. In this role, you will design, build, and productionise intelligent capabilities that combine agentic generative AI systems with conventional ML pipelines across identity resolution, audience modeling, attribution, and optimisation workflows. You will work with massive datasets, develop pipelines that transform raw signals into actionable features, and partner closely with product, data science, and platform teams. This position sits at the intersection of large-scale data platforms, applied machine learning, and generative AI — building capabilities that make Roku's advertising data more actionable for internal teams and advertisers. You will be responsible for building core generative AI platform features for Roku Advertising while also applying traditional ML techniques to improve accuracy, automation, and decisioning across the ecosystem.

How will I use AI at Roku?

At Roku, AI agents do most of the keystroke-level coding. Engineers act as the technical leads for those agents — setting context, planning work, verifying outputs, and recovering when the agent goes off course. If you are excited about working this way, we want to talk to you.

At Roku, we're embracing AI as a powerful tool to amplify human creativity, accelerate innovation, and deliver better results for our customers. Across teams, we're exploring how AI can enhance our work — and we're only looking for people who are curious, adaptable, and excited to grow alongside these technologies.

What are the responsibilities of the role?

• Develop scalable and effective ML models and data pipelines to power identity, graph enrichment, audience segmentation, and campaign performance intelligence
• Design and run experiments, measure impact, and translate results into product decisions and customer outcomes
• Build and productionize agentic, generative-AI features — including reporting and insights agents, campaign-monitoring agents (with human-in-the-loop controls), and a semantic layer for reliable, natural-language analytics
• Improve automation, reliability, and scalability across model training, feature engineering, deployment, and monitoring frameworks
• Work closely with software and ML engineers to deliver end-to-end solutions from data ingestion through downstream systems
• Partner with senior stakeholders to shape ML and Gen-AI strategy for Ad Data Activation — identifying the right problems for AI, ensuring safety, and driving advertiser productivity

What experience would help someone be successful in this role at Roku?

• 5+ years of experience applying ML or data science to real business problems at scale
• Master's degree (or higher) and 4+ years of experience in CS, CE, ML, or a related field
• Strong computer science fundamentals and ability to design efficient algorithms and production systems
• Experience with ML approaches used in large-scale data problems (e.g., regression, classification, tree-based methods, embeddings, sequence models, or basic NLP)
• Strong intuition for defining KPIs, objective functions, and evaluation frameworks for ML-driven systems
• Knowledge of programming languages such as Python, Scala/Java, C/C++, or similar
• Experience working with large-scale data platforms (Spark, Presto/Trino, feature stores, data lakes, etc.)
• Experience building or integrating with LLM/Gen-AI pipelines (prompting, evaluation, retrieval, safety)
• Prior publications, patents, or open-source contributions
• You have built fluency across the agentic engineering toolchain — coding harnesses like Claude Code or Cursor, MCP servers, custom skills, or agent frameworks. And you can describe projects where you shipped real work with these tools. You know how to drive an agent, verify its output, and ramp on an unfamiliar codebase with an agent helping you