Senior Computer Vision / Machine Learning Engineer · Digital Twin Platform

About Woven by Toyota Woven by Toyota contributes to Toyota's transformation into a mobility company once in a hundred years. Inspired by our history of continuing to invent "for someone other than ourselves," our mission is to expand the definition of mobility, extend the ways mobility contributes to society, and continue to challenge the common sense of mobility through innovation that thinks of people.

We operate around four main pillars: AD/ADAS (Automated Driving / Advanced Driver Assistance Systems) technology, Arene—a vehicle software production platform for SDV (Software Defined Vehicles), Woven City—a mobility test course, and Cloud & AI—the digital infrastructure that supports our collaborative foundation. Additionally, essential business functions support the implementation of these teams, working together as one to realize a "zero accident society" and the "mass production of happiness."

=========================================================================

About the Team

Toyota will redefine the meaning of mobility. We will move people, goods, information, and energy, and continue to challenge the common sense of mobility. Woven City serves as a test course to create and test various such mobilities, aiming for the mass production of happiness with the concepts of a human-centric city, a city for demonstration experiments, and an unfinished city.

Woven City brings together diverse colleagues who share a vision for future mobility. Inventors, residents, and visitors will demonstrate various products and services while living in a test course shaped like a city where people, mobility, and social infrastructure collaborate.

As a member of the Social Infrastructure/Common Platform, one of the core teams under the Woven City umbrella, our Digital Twin Platform team is responsible for realizing the Woven City vision: "Expand mobility. Enhance humanity. Engage society." By developing a robust software platform, we enable inventors and business owners to "test" their products in a virtual environment before deploying them to the actual Woven City. Simultaneously, we provide an analytical foundation that supports decision-making based on both operational data collected from the actual city (such as residents' daily lives, transportation dynamics, and energy consumption) and simulation results. Since the Digital Twin Platform is still in its early growth stages, you will have a valuable opportunity to be involved in the process where both the platform and commercial strategies mature, significantly influencing not only "what to build" but also "how to deliver to the market" and "what business models to trial."

For more details regarding Woven City, please see below: https://www.woven-city.global/jpn/

Ideal Candidate

We are looking for a senior-level Computer Vision / Machine Learning Engineer who does not just combine existing APIs and off-the-shelf pipelines, but is deeply involved in model design, training, and tuning. As a senior engineer, you will be expected to lead technical direction and act with high autonomy. Rather than simply executing tasks, you will define approaches to difficult problems, influence the architecture of the entire CV/ML stack, and support the growth of team members. You will tackle computer vision challenges in real-world environments such as logistics hubs, distribution centers, and factories, performing tasks such as detection and tracking of workers, vehicles, and cargo, analysis of space utilization, and identification of productive versus wasteful work through understanding worker behavior. Candidates with research experience or a PhD in related fields are welcome, but a PhD is not mandatory. Even without a PhD, we welcome those with extensive practical experience and an equivalent level of technical curiosity and rigor. As a core member of a small, highly autonomous team, you will directly shape the product direction and its real-world impact through technical judgment.

Responsibilities

• Design, train, and optimize deep learning models for object detection, multi-object tracking, classification, and action recognition for industrial environments.
• Build and operate real-time computer vision inference pipelines and analytical systems for logistics hubs, distribution centers, and factory operations.
• Lead dataset strategies, annotation standards, model evaluation, and experimental design to continuously improve the performance and reliability of CV/ML models.
• Define and drive the technical roadmap, architecture, and integration strategies for CV/ML systems within the broader platform ecosystem.
• Establish engineering best practices and nurture/mentor engineers through technical leadership, design reviews, and the introduction of the latest CV/ML technologies.

Requirements

• A Bachelor's degree or higher in a related field with 7+ years of practical experience in CV/ML engineering, OR a PhD in a related field with 5+ years of applied practical experience.
• Practical experience in model design, training, and hyperparameter tuning, with proficiency in PyTorch or TensorFlow.
• Experience leading the development of scalable Perception pipelines, including state-of-the-art object detection and multi-object tracking systems. A deep understanding of model trade-offs, data quality, evaluation methodologies, and constraints in real-world deployment is required.
• A proven track record of delivering and operating complex CV/ML systems end-to-end in production environments.
• A proven track record of leading technical projects, driving architectural decisions, and influencing engineering direction.
• Sufficient understanding of classical computer vision geometry (camera intrinsic/extrinsic parameters, calibration, and the basics of 3D reconstruction).
• Excellent communication skills in English (reading, writing, and speaking).

Preferred Qualifications

• PhD in Computer Vision, Machine Learning, or a related field.
• Experience in human pose estimation, skeleton-based action recognition, or temporal activity recognition in video.
• Knowledge of edge inference optimization (TensorRT, ONNX export, quantization, pruning, etc.).
• Experience with 3D sensing modalities such as LiDAR, depth cameras, multi-camera calibration, and sensor fusion.
• Professional experience in industrial sectors such as logistics, warehousing, or manufacturing.
• Experience in space utilization analysis, occupancy estimation, or pedestrian/vehicle flow analysis.
• Experience building video annotation pipelines or active learning loops.
• A track record of publishing papers at top-tier international CV/ML conferences (e.g., CVPR, ECCV, ICCV, NeurIPS, etc.).
• Experience in Vision-Language Model (VLM) or agent-based system design.
• Practical experience with cloud platforms (AWS, GCP) for ML workloads.

========================================================================= Notes ・Typically, all interviews are conducted via Google Meet. ・Job postings are currently available in both English and Japanese. Please apply to only one of them. ・When applying, we request that you submit an English resume if possible. Please note that if you submit a Japanese resume for any reason, we may request an English resume during the selection process depending on the position.

Compensation & Benefits ・Salary commensurate with industry standards based on experience. ・Working Hours: Flexible working hours. ・Annual Paid Leave: 20 days per year (number of days in the first year varies depending on the month of joining). ・Sick Leave: 6 days per year (number of days in the first year varies depending on the month of joining). ・Holidays: Saturdays, Sundays, public holidays, and other days designated by the company. ・Social Insurance: Health insurance, Welfare Pension Insurance, Industrial Accident Compensation Insurance, Employment Insurance, Nursing Care Insurance. ・Housing Allowance. ・Retirement Allowance System. ・Rental Car Support. ・Internal Training Programs (Software learning, Language learning).

Our Commitment ・We provide equal employment opportunities and respect diversity. ・Personal information provided will be used only for recruitment and onboarding procedures. For details, please refer to our Privacy Policy.