Product Manager, Multimodal Data Pipelines, AI Foundations
San Jose, CA, USA · Manager · Full-time · Posted Oct 20, 2026
- Experience
- 5 years in product management or a related technical role; 2 years taking technical products from conception to launch. Preferred: 2 years in software development or engineering; 1 year in technical leadership.
- Education
- Bachelor's degree or equivalent practical experience; preferred Master's degree in a technology or business related field.
About the job
Define the strategy and roadmap for multimodal data pipelines that turn web-scale data into structured data for Google’s AI models, Search, and agentic applications.
What you'll do
- Set the strategy for native multimodal data pipelines based on data needs across Search, LLMs, RAG, and information retrieval.
- Build and manage a roadmap that accelerates model development and AI adoption.
- Lead technical products from conception to launch and align cross-functional teams to resolve execution blockers.
What they ask for
- Bachelor's degree or equivalent practical experience.
- 5 years of experience in product management or a related technical role.
- 2 years of experience taking technical products from conception to launch.
- Experience integrating generative AI tools or LLM interfaces into workflows, and building and scaling developer platforms, APIs/SDKs, or backend infrastructure.
Nice to have
- Master's degree in a technology or business related field; 2 years of experience in software development or engineering; 1 year of experience in technical leadership.
- Experience with data processing using LLMs, Model Development, Model Training, or RAG.
Pay
$163000 - $236000 (USD) base salary
Tools
Large Language Models (LLMs), RAG, Retrieval-Augmented Generation (RAG), Gemini, APIs/SDKs
AI focus
Develop multimodal data pipelines and use AI-based signal processing to prepare data for LLM training, RAG, Search, and agentic applications; experience integrating generative AI tools or LLM interfaces is required.
Model-written summary · the original posting is authoritative