Developing artificial intelligence that enables service robots to collaborate with people and adapt to dynamic human environments for a venture-backed company
Requirements
- Strong background in deep learning frameworks (PyTorch, TensorFlow, JAX)
- Hands-on experience building and deploying ML models robotic manipulation for grasping, manipulation, or dexterous robotics
- Expertise with large multimodal ML models (vision-language-action, tactile sensing)
- Experience with simulation for manipulation (MuJoCo, PyBullet, Isaac, or equivalent)
- Familiarity with SLAM, mapping, and navigation pipelines
- Solid software engineering skills in Python and C++ for ML system integration
- Proven ability to take ML models from research prototype to production deployment
Responsibilities
- Design and implement ML algorithms for robotic manipulation, including grasp planning, dexterous manipulation, and tool use.
- Develop and train large multimodal ML models (vision, tactile, language, action) to enable robust manipulation behaviors.
- Build pipelines for data collection, labeling, and augmentation to support manipulation learning.
- Leverage simulation environments (Isaac Gym, MuJoCo, Omniverse, etc.) for training, evaluation, and transfer to real robots.
- Optimize models for onboard, real-time performance on robotic hardware.
- Collaborate with perception, navigation, and platform teams to integrate manipulation skills into the full robotics stack.
- Analyze robot performance, develop benchmarks, and iterate based on real-world deployments.
Other
- Master’s or PhD in Computer Science, Robotics, Machine Learning, or related field
- 5+ years of experience in applied machine learning, computer vision, or robotics
- Contribute to code reviews, documentation, and best practices for ML/robotics development
- Stay current with state-of-the-art manipulation and embodied AI research, bringing promising ideas into production
- Strong debugging skills for diagnosing ML performance gaps in fielded systems
