Software Engineer, ML Platform

The ML Platform team provides foundational tools and infrastructure used by hundreds of ML engineers across Pinterest, including recommendations, ads, visual search, growth/notifications, trust and safety. We aim to ensure that ML systems are healthy (production-grade quality) and fast (for modelers to iterate upon).

• The ML Data team enables rapid development of features and training/evaluation datasets. We provide a batch and real-time feature store, a central ML Dataset Store, as well as a signal platform with monitoring and governance tools. The team works with Spark, Parquet, Flink streaming, as well as Java online services, and Python for orchestration.
• The ML Training team builds large-scale training, model management/deployment tools, and modeling libraries. We build a Training Compute Platform provisioning GPU and cluster compute, model deployment, and PyTorch ML modeling libraries providing cutting-edge model building blocks. The team works with Python, MySQL, and full-stack UI development, as well as interacting with Kubernetes, Spinnaker, Jenkins.
• The ML Serving team builds our universal engine for large-scale ML model inference in online, offline, and streaming contexts. We also develop services to fetch features and provide visibility/monitoring/observability on deployed ML systems. The team works with C++ online serving and hardware optimization.

What you’ll do:

• Design and build core components of the ML lifecycle
• Work with internal customers, ML engineers and data scientists across Pinterest, to understand and solve development velocity pain points
• Work with many major ML teams in Pinterest to productionize their models, improve their productivity and unblock ML innovations

What we're looking for:

• 2+ years of experience, preferably with technologies used on one of the teams above
• Solid communication to solve problems for customers across teams
• Experience with production ML systems, preferably the end-to-end lifecycle of an ML model
• Ability to drive cross-team projects; Ability to understand our internal customers (ML practitioners), their common usage patterns and pain points

#LI-REMOTE

#LI-AH2