Google
Google

Machine Learning Staff Software Engineer, Search Personalization

RoleMachine Learning
LevelStaff
LocationMountain View, Canada, United States
WorkOn-site
TypeFull-time
Posted2 months ago
Apply now

About the role

  • Design and implement personalized user models to optimize for user happiness, including Neural Deep Retrieval Models, Deep Neural Network Ranking/Scoring models, User/Content Clustering Models, Large Language Models (LLM)-based Retrieval Augmented Generation Models, and more.

  • Build user and content clustering models to enable core personalization and ranking use cases.

  • Enhance model performance and personalization precision/recall through advanced modeling techniques such as transformers, distillation, reward shaping, multi-task learning, neural bandits, etc. and capabilities through feature engineering, automatic parameter tuning, label quality engineering, etc.

  • Scale the model's applications to a multitude of modalities (content, queries, videos and notifications) and use cases (retrieval, ranking, content generation, diversification, etc.).

  • Create next-generation realtime ML models that can capture new user interests and world trends in seconds and scale model training and serving to billions of users.

  • Bachelor’s degree or equivalent practical experience.

  • 8 years of experience with one or more general purpose programming languages including but not limited to: Java, C/C++ or Python.

  • 8 years of experience in software development.

  • 5 years of experience building and deploying recommendation systems models (retrieval, prediction, ranking, embedding) in production and experience building architecture in different modeling domains.

  • 5 years of experience with ML design and ML infrastructure (e.g., model deployment, model evaluation, data processing, debugging, fine tuning).

  • 5 years of experience testing, and launching software products, and 3 years of experience with software design and architecture.

About Google

Mountain View

Headquarters