Principal Software Engineer, ML & Media Infra

Pinterest•Palo Alto, CA

6d•$285,452 - $449,541•Remote

About The Position

The Principal Software Engineer will set technical direction for Pinterest’s ML and media infrastructure, enabling fast, safe, and efficient delivery of AI/ML solutions across Ads and Core products. The role focuses on model training and serving, GPU efficiency, and large-scale ranking systems, while also leading platforms for content acquisition, media ingestion, and media signal processing. The team builds unified data, training, feature, and inference infrastructure and collaborates with Content, Shopping catalog, GenAI, Ads, and Core AI/ML teams. The role is remote with in-person collaboration in the office 1–2 times per quarter, and the employee must live within commuting distance of the Palo Alto, CA or San Francisco, CA office. The engineer will shape reusable interfaces between model training and serving and data and feature infrastructure; lead architectures for distributed training, fine-tuning, distillation, evaluation, inference, and media systems; and improve GPU and system efficiency. They will build reliable, observable platforms, drive architecture and operational decisions across organizations, mentor senior engineers, and use AI-assisted development responsibly while ensuring correctness, quality, and data safeguards. The position is not eligible for relocation assistance.

Requirements

Track record of setting technical strategy and delivering company-wide infrastructure initiatives in ambiguous environments.
Deep expertise in distributed ML systems, including production experience with both large-scale training and online inference.
Strong GPU performance knowledge, such as profiling, distributed execution, kernel and memory optimization, compilation, or quantization.
Experience with AI/ML modeling, recommender systems, Ads ranking, retrieval, feature platforms, or similarly demanding ML workloads.
Experience with content acquisition, such as crawlers, content understanding, or near-real-time data processing.
Experience with search and recommendation systems or Ads serving systems.
Strong systems programming and design skills in C++, Java, or Python.
High ownership and sound judgment in reliability, security, cost, and operational excellence.
Ability to use AI to improve speed and critically evaluate AI-assisted work, with accountability for correctness, quality, and sensitive data.
Bachelor’s degree in Computer Science, Engineering, a related field, or equivalent experience.

Responsibilities

Set the technical vision and roadmap for model training and serving, with reusable interfaces to data and feature infrastructure.
Lead architectures for distributed training, fine-tuning, distillation, evaluation, and high-scale CPU/GPU inference.
Lead architectures for content acquisition, media ingestion, and signal processing.
Improve efficiency across data loading, distributed execution, GPU kernels and memory, compilation, quantization, scheduling, and capacity.
Build reliable, observable platforms with strong quality guarantees and training/serving consistency.
Partner with Ads and Core AI/ML teams to productionize features and models safely at Pinterest scale.
Drive cross-organizational architecture decisions, migrations, and operational standards; mentor senior engineers and raise the engineering bar.
Partner closely with Content, Shopping catalog, and GenAI teams to improve Pinterest’s content ecosystem.
Use AI-assisted development and analysis to accelerate prototyping, performance diagnosis, and validation while maintaining rigorous correctness and data safeguards.