Education Requirements:

• Bachelor's degree or higher

Experience Requirements:

• 3+ years of relevant experience

Other Requirements:

• Understanding of autonomous driving systems (perception, control, etc.), ADAS (ACC, AEB, etc.), and similar systems (e.g., robots)

• Understanding and experience with vehicle sensors (Camera, Radar, GPS, etc.)

• Understanding and experience with vehicle testing

• Understanding and experience with vehicle testing equipment (RT, Vector equipment, etc.)

Responsibilities:

• Development milestone-based evaluation strategy and planning

• On-road test scenario development

• Conducting on-road tests based on scenarios and specifications

• Vehicle-level performance development and performance indicator management

• Building and upgrading on-road test environments

Education Requirements:

• Master's or Ph.D. degree in Computer Science, Electrical Engineering, Mathematics, Statistics, or a related field relevant to machine learning, or equivalent experience

Experience Requirements:

• Minimum of 5 years of relevant experience

Other Requirements:

• Strong background in Linear Algebra, Probability, Signal Processing, and machine learning concepts

• Proficient programming skills in languages such as C/C++, Python, and others

Responsibilities:

• Develop algorithms and automated systems for automatically generating labels using various sensor and video data collected during autonomous driving.

• Dataset and evaluation: Curate high-quality datasets tailored to autonomous driving scenarios and design robust evaluation metrics to accurately assess algorithm performance.

• Active learning: Research techniques for efficiently selecting and labeling high-value data points to improve model performance while minimizing labeling efforts.

• Network architecture search: Explore methods for automatically discovering optimal neural network architectures for generating labels from sensor and video data.

• Transfer/low-shot/long-tail learning: Develop strategies to leverage knowledge from related tasks or domains to address challenges with limited labeled data or class distribution imbalances.