PNNL is seeking a Data Scientist to contribute to cutting-edge research in robotic autonomy, learning-enabled control, and embodied AI, and to help develop new capabilities in optimization, machine learning, and integration.
Requirements
- PhD or MS in Robotics, Computer Science, Applied Mathematics, Electrical Engineering, Mechanical/Aerospace Engineering, or related scientific fields
- 5–10 years of hands-on experience in robot learning, motion planning, navigation, and control using both classical and modern control methods (e.g., MPC, PID, LQR) and modern machine learning techniques (e.g., reinforcement learning, imitation learning, computer vision)
- Experience with physics-simulation frameworks such as MuJoCo, Gazebo, and IsaacSim
- Strong programming skills in Python/Julia/C++, ROS, and machine learning frameworks (e.g., PyTorch, JAX, TensorFlow)
- Experience working with robotic systems such as manipulators, mobile robots, autonomous vehicles, or similar platforms
- Experience using machine learning in cloud environments (such as Google Cloud and AWS) and edge hardware for real-time deployment and scalability
- Experience with modern scientific deep learning methods (e.g., Neural ODEs, PINNs, Operator networks, Hamiltonian and Lagrangian neural networks, graph neural networks)
Responsibilities
- Designs, develops, and implements methods, processes, and systems to analyze diverse data.
- Applies knowledge of statistics, machine learning, advanced mathematics, simulation, software development, and data modeling to integrate and clean data, recognize patterns, address uncertainty, pose questions, and make discoveries from structured and/or unstructured data.
- Produces solutions driven by exploratory data analysis from complex and high-dimensional datasets.
- Designs, develops, and evaluates predictive models and advanced algorithms that lead to optimal value extraction from the data.
- Develops and applies advanced algorithms for motion planning, learning-enabled control, and autonomous decision-making using techniques such as model predictive control (MPC), control barrier functions (CBFs), differentiable predictive control, and reinforcement learning.
- Contributes to software development and applied mathematics research in robotics for scientific applications such as autonomous laboratories.
- Helps design and maintain codebases that support scalable experimentation, dataset acquisition, and integration of ML-enabled control systems in both simulated and physical testbeds.
Other
- BS/BA and 7+ years of relevant work experience -OR- MS/MA and 5+ years of relevant work experience -OR- PhD with 3+ years of relevant experience
- Excellent communication and interpersonal skills are essential for engaging in interdisciplinary research and delivering impactful scientific outcomes
- Ability to mentor junior staff members, post-doctoral researchers, and seek programmatic funding
- Commitment to inclusive research practices, scientific integrity, and collaborative team culture
- Onsite work at PNNL main campus in Richland, WA
