Sr. Machine Learning Engineer, Applied Science

About the role

Pinterest Labs is actively seeking a Senior Machine Learning Engineer to spearhead the Pinterest Canvas initiative, driving the evolution of text-to-image foundation models. In this capacity, you will own the complete product lifecycle, transforming nascent research hypotheses into robust, scalable production features. Your primary mission involves mentoring cross-functional peers and articulating a coherent, long-term generative visual strategy for the platform. This position is open in San Francisco, California, or as a remote role for eligible candidates across the United States. You will be responsible for architecting the next generation of text-to-image pipelines, with a specific focus on optimizing for Canvas deployment in real user environments. A central pillar of this role will be orchestrating the collection and curation of high-fidelity visual training data to fuel model improvement. You will design and implement reinforcement learning from human feedback (RLHF) frameworks to systematically enhance the quality and safety of generated outputs. Leading research-oriented reading groups will fall under your purview, enabling the team to rigorously evaluate academic papers and define future model capabilities. You will ensure rigorous model deployment through experimentation, monitoring, and deep analysis of core user journeys and interactions. Close collaboration with product managers and infrastructure engineers will be essential to guarantee that model behavior aligns with Pinterest's platform standards and scalability demands.

Key facts

What you'll do

• Execute the technical vision for the Pinterest Canvas initiative, translating high-level research into production-grade text-to-image models.
• Architect and refine end-to-end text-to-image pipelines, optimizing for latency, stability, and output quality in a live product context.
• Orchestrate the strategic collection and curation of visual training data to build and maintain high-performance generative models.
• Implement and manage reinforcement learning from human feedback (RLHF) workflows to iteratively improve generative quality and safety.
• Lead internal research reading groups to dissect and assess the latest academic publications and emerging methodologies.
• Shape the long-term generative visual roadmap for Pinterest Labs based on technical feasibility and user impact analysis.
• Rigorously ship models by conducting experimentation, monitoring system performance, and analyzing core user journey metrics.
• Review model outputs and synthesize user feedback to drive iterative enhancements in quality, safety, and real-world performance.
• Partner with infrastructure teams to ensure model serving is reliable, scalable, and aligned with Pinterest's technical stack requirements.
• Collaborate with product specialists to integrate embeddings and multimodal features that effectively condition the Canvas model.
• Enable the effective use of Pinterest's massive visual-text dataset for downstream product capabilities and feature development.
• Mentor junior researchers and interns, fostering a high-performing culture within the Pinterest Labs environment.
• Evaluate and recommend new tools and frameworks that enhance the efficiency and effectiveness of the modeling workflow.
• Ensure all model implementations adhere to platform requirements and contribute to a cohesive user experience.

Requirements

• Demonstrate hands-on experience building generative computer vision systems, with a strong emphasis on diffusion model architectures.
• Hold a Master of Science or PhD in Machine Learning, Computer Science, or a closely related quantitative field.
• Bring a minimum of five years of professional experience building computer vision systems within an industry setting.
• Exhibit proficiency in designing, training, and deploying machine learning models at scale.
• Show a deep understanding of the end-to-end machine learning lifecycle, from data collection through deployment and monitoring.
• Possess strong software engineering practices, including version control, testing, and code review.
• Communicate complex technical concepts effectively to both technical and non-technical stakeholders.
• Work effectively in a fast-paced, ambiguous environment while managing multiple priorities and deadlines.
• Collaborate closely with cross-functional teams to solve large-scale, real-world engineering problems.
• Commit to upholding the highest standards of safety, ethics, and quality in generative AI applications.

Nice to have

• Have published work at top-tier machine learning conferences.
• Utilize modern AI coding assistants such as Cursor, Copilot, or Codex in your development workflow.
• Leverage large language model-powered tools for documentation search, experiment analysis, SQL exploration, and engineering workflow optimization.

Practical notes

Application Details

To apply, please follow the instructions on the official Pinterest careers website. We encourage you to