Staff Computer Vision & Multimodal ML Engineer

Pinterest; San Francisco (CA); Hybrid; USD 189,308 - 389,753; Full time

Job summary

Pinterest is seeking researchers and engineers to advance a cutting-edge visual encoder within the Labs’ Visual Modeling team. You will work with large visual-text datasets to train multimodal representations powering recommender systems and visual language features.

Responsibilities include prototyping state-of-the-art encoders, handling billion-scale datasets with GPU computing, and publishing results while mentoring interns. A strong ML background and publications are highly valued.

Qualifications

• Experience building and training computer vision models.
• Experience with multimodal representations and visual language modeling.
• Track record of research contributions or shipping ML models to production.
• Hands-on experience with large-scale model training and modern DL frameworks.

Responsibilities

• Prototype state-of-the-art visual encoders for recommender systems.
• Experiment with billion-scale datasets using GPU computing.
• Build flexible visual reasoning tools and generation models.
• Read papers, discuss research, and shape Pinterest's visual generative strategy.
• Mentor junior researchers and researchers interns within Labs.

Skills

Computer vision

Multimodal learning

Research experience

Collaboration

Education

MS/PhD in ML or related field

Tools

PyTorch

Cursor/Copilot/Codex

Job description

Pinterest is seeking researchers and engineers to advance a cutting-edge visual encoder within the Labs’ Visual Modeling team. You will work with large visual-text datasets to train multimodal representations powering recommender systems and visual language features.

Responsibilities include prototyping state-of-the-art encoders, handling billion-scale datasets with GPU computing, and publishing results while mentoring interns. A strong ML background and publications are highly valued.