Senior Deep Learning / Machine Learning Engineer, Perception & Prediction

Key Skills

Java Programming, Python Programming, C++, C Programming, SQL, Cloud computing, Scala Programming, Machine learning techniques, Data science techniques, MATLAB Programming, PyTorch, TensorFlow, R Programming

Job Description

Woven Planet is building the safest mobility in the world. A subsidiary of Toyota, Woven Planet innovates and invests in new technologies, software, and business models that transform how we live, work and move. With a focus on automated driving, smart cities, robotics and more, we build on Toyota's legacy of trust and safety to deliver mobility solutions for all.

For nearly a century, Toyota has been delivering products and services that improve lives. Automation that originated to increase the efficiency of daily activities has evolved into the safe, reliable, connected automobiles we enjoy and depend on today. Now, we are looking to the next 100 years and to extending that dream for a better life for all people. At Woven Planet we strive to build a safer, happier, more sustainable world.

Our unique global culture weaves modern Silicon Valley innovation and time-tested Japanese quality craftsmanship. The complementary strengths enable us to optimize safety, advance clean energy, elevate well-being, and improve how people live, work, and play. We envision a human-centered future where world-class technology solutions expand global access to mobility, amplify the capabilities of drivers, and empower humanity to thrive.

About the Organization

Woven Planet is developing automated driving technology using a data-driven approach. We’re building products at autonomy levels 2-4 to drive both near- and long-term improvements to mobility for all. Woven Planet has the backing of one of the world’s largest automakers, the talent to deliver on our goal, and a built in path to product and revenue—a combination rarely seen in the mobility industry. We’re looking for doers and creative problem solvers with a passion for improving lives.

Each member of our diverse and talented group of software and hardware engineers has the opportunity to make a meaningful impact on our technology and products. Our growing team works in brand new garages and labs in Palo Alto, tests AVs at our dedicated test track in Silicon Valley, and explores the industry’s most compelling research problems at our office in London. With support from our Woven Planet colleagues in Tokyo, our work to improve the future of mobility spans the globe.

You will be interacting on a daily basis with other software engineers and researchers to tackle some of the most challenging problems in AI, robotics, and computer vision. We work on a diverse set of problems ranging from solving optimization problems in 3D geometric computer vision, to minimizing latency on hardware accelerators, to designing novel neural network architectures, to applying and advancing the state-of-the-art of machine learning for Perception, Prediction, and Motion Planning.

The team is looking for deep learning experts to drive the research and deployment of the next generation of deep learning models for Autonomous driving. The ideal candidate would have published deep learning research in top-tier conferences such as NeurIPs, CVPR, ICRA or ICLR, built and deployed real-world deep learning products and worked in a fast-paced environment along with other highly talented engineers. We recognize the unique capabilities each team member can bring, though and encourage applicants to reach out even if they do not match all of the characteristics described below. 

Responsibilities:

• Work in a small, high-velocity team of engineers
• Collaborate closely with stakeholders in Perception and other teams (Compute, Motion Planning, etc.) to define interfaces and requirements for the Perception stack
• Investigate and develop ML models for calibration, mapping, detection, segmentation, classification, sensor fusion, tracking and prediction based on a heterogeneous sensor architecture including cameras, RADAR and LiDAR
• Analyze and optimize deep learned models for deployment on state-of-the-art edge-compute architectures with significant neural network acceleration capabilities
• Set up pipelines and develop strategies for mining large-scale fleet data to train key ML models
• Develop infrastructure and tooling to introspect models and evaluate performance 
• Deploy models in a production self-driving vehicle environment, validating and optimizing various performance measures such as latency and accuracy
• Advance the state-of-the-art and represent Level 5 by publishing at top-tier conferences or online blog articles

Experience:

• MS or PhD (preferred) in Computer Vision, Machine Learning, Robotics, or other quantitative field or relevant work experience
• 3+ years experience as Perception Engineer
• Hands-on experience implementing state-of-the-art deep learned ML solutions
• Strong understanding of ML workflows: data sampling and curation, pre-processing, model training, ablation studies, evaluation, deployment, inference optimization. 
• Expertise in building deep learning models and systems with Python (PyTorch/TensorFlow) and C++
• Strong attention to detail and analytical skills
• Excellent communication and technical writing skills (design documents, ML reports)
• (Nice to have) Experience with quantization-aware training and low-precision inference
• (Nice to have) Experience targeting power-efficient, edge-compute architectures
• (Nice to have) Experience working on self-driving problems (Perception, Prediction, Mapping, Localization, Planning, Simulation)
• (Nice to have) Expertise (MS or PhD-level) in probabilistic modeling, high performance compute, dynamical systems, or computational geometry