Google's Traffic AI team builds core machine learning models for accurate, real-time travel time and traffic information for over a billion users worldwide, impacting Google Maps and other products. The role aims to solve complex, real-world user problems related to navigation and ETA accuracy.
Requirements
- 5 years of experience with software development with Python and C++.
- 5 years of experience with machine learning, artificial intelligence, or a related field.
- Experience with common machine learning frameworks such as TensorFlow or PyTorch.
- Proficiency in Python, TensorFlow, and C++.
- Expertise in applied machine learning, including neural network architecture, model tuning, and loss function design.
- Ability to translate abstract product problems into concrete ML formulations.
- Experience with large-scale data processing, data analytics, and end-to-end debugging in a production environment.
Responsibilities
- Take end-to-end ownership of user-facing issues, translating ambiguous problems into clear machine learning formulations.
- Drive the full lifecycle of model development, from data processing and featurization to designing and tuning neural network architectures.
- Search into large-scale, real-time production data to diagnose issues, identify data quality problems, and debug the entire data-to-model pipeline.
- Design and execute robust offline and online experiments to validate model performance and its impact on product metrics.
- Collaborate closely with Research, Infrastructure, and Product teams to innovate and deliver solutions that improve navigation for millions of users.
Other
- Bachelor’s degree or equivalent practical experience.
- Master’s degree or PhD in Engineering, Computer Science, or a related technical field.
- Excellent communication skills and a passion for solving user-facing problems.
- Versatile, display leadership qualities and be enthusiastic to take on new problems across the full-stack.
