Introduction: Join our innovative team as an Expert Scientist, Dexterous Manipulation in Vancouver, BC. This role offers an opportunity to work on cutting-edge robotic manipulation tasks, leveraging state-of-the-art AI and machine learning techniques.
About the Company: We are a trailblazing company specializing in humanoid robots and novel control systems. Our focus is on integrating symbolic logic with data-driven models to achieve human-like intelligence in general-purpose robots. Established in 2018, we have rapidly grown and earned recognition as a top startup, moving into a state-of-the-art facility in 2023.
About the Role: As an Expert Scientist, Dexterous Manipulation, you will lead the development and implementation of advanced reinforcement learning (RL) and imitation learning (IL) algorithms. Your work will involve creating training and data collection systems, evaluating these algorithms in both simulated and real-world environments, and utilizing novel haptic and proprioceptive sensing techniques with our proprietary robots.
What We Can Offer You:
- Competitive salary and equity stakes
- Comprehensive health coverage
- Generous paid time off
- Access to cutting-edge work facilities
- Flexible worksite options based on role
Key Responsibilities:
- Develop and enhance RL and IL algorithms, ensuring practical application success
- Stay updated with the latest RL/IL techniques and their robotics applications
- Lead research initiatives and communicate findings to the broader ML team
- Optimize RL/IL training and data collection pipelines for physical robots
- Collaborate within a diverse team to innovate and troubleshoot algorithmic errors
Qualifications:
- Ph.D. in Machine Learning, Computer Science, Applied Mathematics, or equivalent practical experience in RL and IL
- 5+ years in robotic manipulation tasks, both in simulation and real-world settings
- 5+ years in applying RL and IL methods, focusing on real-world robotics
- 4+ years in developing and optimizing large-batch parallel simulations for RL
- Proven expertise in continual learning and sim-to-real transfer
- Experience in transitioning ML research and models into production
- Active involvement in integrating ML models into a robotics platform
- A track record of publishing research in esteemed AI conferences such as ICRA, IROS, and CORL
Skills:
- Proficiency in Python 3.8 or later
- Working knowledge of PyTorch and/or TensorFlow
- Familiarity with ROS2
- Expertise in RL principles and applications
- Experience with Atlassian tools; Jira, Confluence, or equivalent like GitLab
Traits:
- Positive attitude and willingness to tackle complex problems
- Strong leadership skills in organizing R&D projects
- Tenacity and positivity in facing new challenges
- Attention to detail in resolving performance issues
- Enthusiasm for advancing human-like intelligence in machines
- Ability to multitask and prioritize in a fast-paced environment
Relevant Keywords: AI/Automation, machine learning, robotic manipulation, reinforcement learning, imitation learning, dexterous manipulation
Join us and be part of a team pushing the boundaries of AI and robotics. Your expertise will help shape the future of intelligent machines. This position is hybrid, based in Vancouver, Canada. A relocation reimbursement and work visa can be provided.
