Staff Software Engineer - Ads Serving Platform at Pinterest

We are looking for multiple staff engineers to initiate, design, and build the next-gen version of key infra components in our monetization ecosystem

What we're looking for:

• 6+ years of relevant industry experience with distributed systems, transactional datastores, and systems programming.
• Experience in building and owning large scale high performance infrastructure powering ads, recommendation, search, or other consumer facing applications.
• Experience solving end-user problems and envisioning solutions to improve their productivity.
• Proficiency in Java, C++, or Python.

Pinterest is one of the fastest growing online advertising platforms and our continued success depends on rapidly scaling our core revenue-generating systems. Specifically, we need 10X the scale of our campaign management, ad delivery, and machine learning platforms, while enabling developers inside Pinterest and external advertisers to build and iterate rapidly on new features. We are looking for multiple staff engineers to initiate, design, and build the next-gen version of key infra components in our monetization ecosystem, such as modernizing an end-to-end ML platform serving over hundreds of use cases making billions of predictions per second, and redesigning our catalog ingestion and ads delivery systems to become one of the leading advertising platforms in the world. These roles are exciting, because you will be able to lean on your deep infra knowledge to redesign systems to handle a much bigger scale, while also having the chance to work with very experienced engineers and cross-functional partners.

What you'll do:

• Re-architect core catalog, ads indexing and serving infrastructure to achieve greater scalability, freshness, performance, and reliability, using data storage, streaming processing, and information retrieval technologies such as MySQL, TiDB, Flink, and HNSW.
• Modernize the ML ecosystem for the entire Pinterest Ads product, replacing a hodgepodge of out-of-date ML models with a unified, modern, and privacy-first ML stack with Pytorch, Spark, Iceberg, and GPU based serving.
• Collaborate with cross-functional teams to define problems and drive solutions.
• Work with a strong team of engineers and provide technical guidance and mentorship.