Staff/Principal Engineer - Perception Capabilities

Motional•Pittsburgh, PA

15d•$200,000 - $285,000

About The Position

As a seasoned expert in Perception Capabilities, you'll field cutting-edge software products and solutions using state-of-the-art computer vision and deep learning. Your breadth of experience spans critical problem areas like detection, classification, segmentation, and tracking, along with essential supporting functions such as data engines, labeling policies, training pipelines, model optimization, simulation, and testing. You're driven by seeing your work significantly impact the product, taking initiative and owning outcomes. You're eager to apply these skills to the perception stack of self-driving vehicles. You understand the unique challenges of safety-critical, open-world applications and are committed to ensuring engineering excellence and rigorous evaluation. You'll devise creative solutions to complex problems, collaborating with interdisciplinary, capability-focused teams across the organization to design, implement, and validate end-to-end solutions.

Requirements

Master's or Ph.D. in Machine Learning, Computer Science, Robotics, or a related field; or equivalent industry experience
Strong leadership skills in executing large, complex technical initiatives
Track record of successful interdisciplinary work and product-focused engineering
Broad understanding of Deep Learning algorithms, architectures, and applications
Experience designing, training, and analyzing neural networks for applications relevant to perception
Fluency in Python, including common libraries for deep learning (PyTorch), scientific computing, and data analysis and visualization
Advanced knowledge of software engineering principles, including software design, source control management, build processes, code reviews, and testing methods
Excellent communication and interpersonal skills
Experience mentoring and leading others

Nice To Haves

Experience shipping a “real” product and balancing ongoing support with next-generation innovation
Experience developing safety-critical products or components
Publications in relevant conferences (CVPR, ICML, NeurIPS, etc.)
Strong programming skills in C++ and/or CUDA programming
Experience with TensorRT and model quantization

Responsibilities

Lead the development and implementation of solutions that span across perception components and directly interface with downstream modules in the autonomy system.
Analyze the performance of the perception subsystem to proactively identify potential limitations with respect to a set of system-level capabilities.
Document potential hazards and lead the cross-team efforts to eliminate or mitigate them.
Collaborate on innovative solutions to address “long tail” challenges. For example, developing techniques for synthetic data generation, advanced data mining, or leveraging language embeddings in our onboard models.
Depending on your background and interests, you may also help optimize and deploy models to the vehicle, ensuring they operate within the constraints of our onboard hardware and meet our functional requirements and interface specifications.
Interface with systems and safety engineers, test engineers, and data analysts to ensure appropriate requirements, test coverage, and performance metrics