Data Scientist - Perception

Responsibilities

• Design and develop our strategy for data development and model evaluation metrics to ensure safe and effective autonomous driving, including defining objective success criteria for ML model deployment 
• Support data-centric AI by: exploring how data impacts ML model performance; guide data labeling strategies; build frameworks for measuring and improving on label quality; develop mining and active learning strategies.
• Define key performance metrics that will help Perception engineers to directly measure the impact of the new features they are developing on the end-to-end driving behavior. 
• Stay up-to-date with the latest developments in machine learning validation techniques and best practices for perception in autonomous vehicles. 

Qualifications

• Bachelor's, Master's or PhD in CS, statistics, mathematics or relevant fields, including statistics (bayesian, parametric statistics, non-parametric, experiment design) and their python-based frameworks (numpy, scipy)
• Experience designing and building evaluation systems for classical and deep-learning models, including linking standard model metrics to end-to-end system behavior and key business metrics
• Experience building and presenting summary metrics to key stakeholders, including advanced knowledge of data visualization techniques
• Experience handling big data, including exposure to distributed data and compute (e.g. Hadoop, Spark, Hive, MapReduce, Databricks), and traditional data query & manipulation (e.g. SQL, NoSQL, Databases)
• Experience adhering to software best practices, e.g. unit testing, DRY, abstraction, modularity, clean interfaces

Bonus Qualifications

• Experience building and/or evaluating ML models that operate on raw sensor data (detection, classification, tracking), including ethical AI validation
• Experience in the automotive industry or autonomous driving systems
• Experience with Databricks 
• Knowledge of perception systems or computer vision