Machine Learning Engineer II, Visual AI

What you'll do

• Prototype new model architectures for Pinterest vision-language models (VLMs).
• Fine-tune open-source LLM models to improve visual perception and tool-using capabilities.
• Develop specialized evaluation benchmarks for vision-centric capabilities like fashion style recommendations.
• Collect relevant visual training data for RLHF and targeted fine-tuning of Pinterest Canvas.
• Contribute to the company's overall visual generative strategy through research and brainstorming.
• Publish research findings via conference submissions, papers, and blog posts.
• Mentor junior researchers and interns within the Pinterest Labs organization.

What we're looking for

• Must have an M.S. or PhD in Machine Learning, Computer Science, or a related field.
• Must have at least 2 years of industry experience in computer vision.
• Experience working with generative computer vision models, including visual encoders and LLMs.
• Experience finetuning open-source LLM models to improve visual perception and tool-using capabilities.
• Ability to develop evaluation benchmarks for vision-centric capabilities like fashion style recommendations.
• Ability to publish research at conferences, via paper submissions, or through blog posts.
• Preferred experience with publications at top ML conferences.
• Experience using AI coding assistants and LLM-powered productivity tools for engineering workflows.