Senior Software Engineer II - Ads Quality

Role overview A senior engineer is needed on the Ads Quality team to lead ML infrastructure for an advertising platform spanning the full serving funnel—retrieval, prediction, pricing, and auctions. The work focuses on building scalable training and serving systems that enable next-generation modeling, including real-time sequential foundation models and LLM-powered capabilities. This is a remote, senior IC role with technical leadership responsibilities, collaborating across ML, data, and platform teams.

Responsibilities - Design and deploy ML infrastructure for ads training and serving, partnering with platform ML teams on shared foundations. - Profile and optimize training and serving bottlenecks to improve throughput, cost, latency, GPU utilization, and tail latency. - Build low-latency real-time feature pipelines that compute, join, and serve fresh features on every ad request. - Diagnose instabilities and operational gaps to deliver high availability for production systems. - Set technical direction for data, feature processing, and model serving solutions, including bespoke foundation models.

Requirements - 3+ years of industry experience building ML infrastructure or applying ML to large-scale real-world problems. - Deep expertise building and scaling production ML training and serving systems, with strong MLOps knowledge. - Strong programming skills (e.g., Python, Go) plus data tooling (SQL, Spark, Pandas) and ML frameworks (PyTorch, TensorFlow, XGBoost). - Strong analytical and problem-solving abilities. - Excellent verbal and written communication with cross-functional stakeholders at all levels.

Nice to have - 5+ years tech-leading a team building ML infrastructure on large datasets. - Experience optimizing training pipelines and serving architectures for efficiency and scalability. - Hands-on experience with streaming systems (Kafka, Flink), online feature stores, and low-latency retrieval. - Experience with sequential modeling, transformers, generative retrieval, or large-scale recommendation systems. - Experience serving large sequential or foundation models in low-latency, high-QPS production. - Background in digital advertising platforms. - Familiarity with LLM integrations and prompt engineering.

Benefits and work setup - Remote-first work model. - Location-based base pay ranges for US candidates: CA/NY/CT/NJ $230,000–$242,500; WA $220,000–$232,000; selected states $211,000–$222,500; all other US states $192,000–$202,500. - New hire equity grant plus annual refresh grants.