Senior AI Research Scientist, Model-based RL

This is a fully remote position, open to applicants in United Kingdom.

📋 Description

• Design, develop, and assess model-based reinforcement learning agents, including controllers based on MPC and MPPI planning.

• Create software prototypes intended for implementation on actual industrial control systems.

• Develop learned dynamics and world models that are applicable across various systems.

• Construct training pipelines that encompass pretraining, curriculum learning, active/adversarial learning, and fine-tuning.

• Investigate and execute safe reinforcement learning, constrained control, scenario planning, and Bayesian reinforcement learning methodologies.

• Document and communicate research results and advancements both internally and externally, through verbal and written means.

• Engage in and coordinate collaborative research initiatives.

• Collaborate with external partners to convert research into production-ready outcomes.

• Mentor and support Research Engineers in applying research to industrial applications.

• Establish new research pathways, convert research findings into practical applications, and oversee the development and implementation of a research domain or significant project.

⛳️ Requirements

• PhD in a relevant technical discipline or equivalent practical experience.

• Solid foundation in model-based reinforcement learning.

• Understanding of planning algorithms, world models/learned dynamics surrogates, reinforcement learning and deep learning, control theory, or safe/constrained reinforcement learning.

• Minimum of 2 years of research experience post-PhD or at least 5 years of research experience post-Master’s degree.

• Extensive research background in Model-Based, Model-Free, and Safe RL and Control Theory, with a particular emphasis on model-based techniques.

• Practical experience in constructing and assessing agents using simulators and bridging the sim-to-real gap.

• Alignment with Phaidra's core values: Agency, Velocity, Craft, and Truth.

• Proficient in Python and PyTorch, including experience with vectorized/differentiable simulators and distributed computing.

• Established history of publications in reinforcement learning, control, or a related field.

• Legally authorized to work in the United Kingdom; employment sponsorship is not available.

• Candidates who progress beyond the initial screening will be required to sign a Non-Disclosure Agreement (NDA).

🏝️ Benefits

• 100% remote work.

• Competitive salary and meaningful equity options.

• Significant responsibilities and opportunities for professional growth.

• Functional training, customer immersion, and development training.

• Comprehensive medical, dental, and vision insurance (specific benefits may vary by region).

• Unlimited paid time off, with a minimum requirement of 20 days annually.

• Paid parental leave (specific benefits may vary by region).

• Flexible stipends for workspace, well-being, and ongoing professional development.

• Company-provided MacBook.

• Virtual team-building activities and social events.