Research Scientist

About the Role

Hedra is assembling a premier Physical AI research team focused on advancing action-conditioned world models and generative AI applied to physical systems. This role offers the opportunity to conduct pioneering research at the intersection of generative modeling, embodied AI, and tangible real-world challenges in partnership with industry collaborators. Researchers benefit from access to extensive computing resources, autonomy in selecting impactful research topics, and a direct pathway to publishing in prominent scientific conferences and journals. The team collaborates closely with esteemed academic leaders, including Fei-Fei Li and the Stanford Vision & Learning Lab.

Key Responsibilities

• Lead and shape research directions involving action-conditioned world models, generative modeling, and embodied physical AI systems
• Innovate by designing new architectures, training objectives, and evaluation methods for vision-language models, vision-language-action models, and world models
• Steer research projects with the aim of securing publications in top-tier scientific outlets
• Collaborate with industrial partners to apply research to practical physical AI scenarios
• Provide mentorship to research engineers and work cross-functionally to translate research into deployable solutions
• Keep abreast of the latest scientific developments, integrating relevant new findings to guide research agendas
• Enhance Hedra’s research ecosystem and contribute to its reputation in the scientific community

Required Qualifications

• Doctorate (PhD) in Machine Learning, Computer Science, Robotics, or related disciplines with research published in leading ML or robotics conferences
• Expertise in generative modeling, world models, or vision-language/action models
• Proven publication record at venues such as NeurIPS, ICML, ICLR, CVPR, CoRL, or similar
• Experience managing large-scale deep learning model training and utilizing modern infrastructure
• Ability to independently initiate and drive research projects through to publication
• Preferred experience includes embodied AI, robotic manipulation, or sim-to-real transfer techniques
• Knowledge of reinforcement learning from human feedback (RLHF), direct preference optimization (DPO), or model alignment strategies considered advantageous
• Strong collaboration and communication skills bridging research and applied implementation teams

Compensation and Benefits

• Competitive salary ranging between $200,000 and $325,000 annually, inclusive of equity options
• 401(k) retirement plan offered without employer matching
• Healthcare coverage including Silver PPO Medical, vision, and dental plans
• Onsite provided lunch and snacks

Additional Information

Candidates are encouraged to apply even if they do not meet every listed requirement, as diverse experiences and unique skill sets are highly valued within our team.