At a glance
Develop and own object detection and tracking algorithms for robots, 5+ years experience
What you will do
- Own architecture and roadmap for object perception systems
- Design, develop, deploy multi-object detection and 6-DoF pose estimation algorithms
- Optimize deep neural networks for embedded systems and collaborate cross-functionally
What they ask for
- 5+ years deploying ML-based object detection on mobile robots
- Master's or PhD in AI, Robotics, Computer Science or related
- Proficiency with deep CNNs, multi-object tracking, PyTorch, OpenCV
Nice to have
- Experience with YOLO, Faster/Mask R-CNN
- Experience in 3 or 6 DoF pose estimation ML models
- Technical leadership experience
Summarised by Banseog from the company's official posting. The official posting is the reference for conditions and details.
What the posting states
Only what the company's official posting says, in its own words.
- Experience
- 5+ yearsRequired
…ties with product requirements Requirements: 5+ years of experience deploying machine learning-based obje
- Education
- Master's · PhDRequired
MASTER/PHD:Artificial Intelligence, Robotics, Computer Science, or a related discipline, with a stron
- Level
- StaffMentioned
Staff AI Engineer, Perception
- Pay
- $207,000 – $323,000Mentioned
Anticipated Salary Range $207,000 — $323,000
- Bonus
- MentionedPreferred
Annual Discretionary Bonus: Provided for eligible roles.
- Equity
- MentionedMentioned
Equity: Company stock options.
- Work authorization
- MentionedRequired
Applicants must have current authorization to work in the United States.
- Relocation
- MentionedMentioned
Relocation Assistance: Provided for eligible roles.
- Benefits
- MentionedMentioned
Benefit Start Date: Eligible for benefits on your fi…
Not stated in the posting: Visa sponsorship · Language. Not stated does not mean not offered. Check the official posting.
Skills in the posting
Similar jobs
See all jobs →Same role · AI & ML
Banseog checks official company careers pages every day. A job is marked closed only when the official page returns a removal (404 or 410).