ML Platform Engineer to join a team at the intersection of sysops, systems programming, architecture, and large-scale deployments. The platform underpins a real-time bidding agent and ML training system that drive $100M+ in annual revenue. Focus on building a Kubernetes + Ray backend for model training pipelines, and defining the next generation of the training and serving stack.

Lead the team that builds the serving and deployment infrastructure for all AI/ML models at Pinterest, including ultra-high-performance C++ model inference engines and production GenAI & LLM model inference stacks.