Requirements

• Bachelor’s degree or equivalent practical experience.
• 1 year of experience with software development in one or more programming languages (e.g., Python, C, C++, Java, JavaScript).
• 1 year of experience with data structures or algorithms.
• Experience integrating generative AI tools or large language model (LLM) interfaces into workflows.

Nice to haves

• Experience with Deep Learning, Information Retrieval, and Multimodal Machine Learning.

What you'll be doing

• Develop universal vector representations (embeddings) of products and related items such as queries, brands, and categories.
• Build two-tower (dual-encoder) Transformer models using a shared encoder starting from a pretrained Gemini model.
• Train models on large datasets with user clicks and relevance labels distilled from a larger model.
• Run training using shared Gemini infrastructure, encode queries online with low latency, and embed/index product listings offline for fast nearest-neighbor vector search.
• Optimize multimodal Gemini embedding models that encode product thumbnails alongside titles and descriptions to improve recall.

Perks and benefits

• Work on projects critical to Google’s needs with opportunities to switch teams and projects as the business grows.
• Join a team responsible for building shopping experiences serving billions of user interactions per day.
• Be part of an equal opportunity and affirmative action employer committed to a culture of belonging and diversity.