Member of Technical Staff - AI Research

Atoms•San Francisco, CA
•$273,000 - $321,000•Onsite

About The Position

Atoms is building Physical AI systems—real-world robots for industries including food, mining, and transport. The company integrates hardware, software, AI, operations, manufacturing, and real estate, deploying its systems in real environments and improving them through operation and learning. As a Member of Technical Staff focused on AI Research, you will define and develop foundational machine learning architectures and learning methods for Atoms' physical AI systems. This research-oriented, deeply technical individual contributor role addresses open-ended problems across foundation models, multimodal learning, world models, computer vision, robotics, and embodied intelligence. The work aims to develop AI systems that build sophisticated representations of the physical world, understand how environments evolve, reason about possible outcomes, and enable intelligent action. The role has significant influence over Atoms' long-term AI strategy and technical architecture. The position is based in Atoms' San Francisco office. Office-based teams work onsite five days a week. The stated base salary range is $273,000–$321,000 per year; actual compensation may vary based on experience, skills, and qualifications. Equity awards may also be available. Benefits listed apply to USA full-time exempt employees and are subject to change.

Requirements

  • Deep expertise in machine learning, with demonstrated experience developing novel models, architectures, algorithms, or learning methodologies.
  • Strong understanding of modern deep learning architectures and foundational machine learning principles.
  • Research experience in one or more areas such as foundation models, multimodal learning, representation learning, computer vision, world models, robotics, or embodied AI.
  • Experience designing and executing research experiments, evaluating hypotheses, and translating findings into functioning models or systems.
  • Strong understanding of large-scale model training, including the relationship between architecture, data, compute, and model performance.
  • Experience with one or more of self-supervised learning, generative modeling, predictive learning, reinforcement learning, or sequential modeling.
  • Strong mathematical and theoretical foundations in machine learning, optimization, statistics, or related disciplines.
  • Hands-on proficiency in Python, PyTorch, or comparable machine learning frameworks.
  • Ability to independently investigate ambiguous research problems and develop novel technical approaches.
  • A track record of meaningful research or technical contributions through deployed systems, novel architectures, publications, open-source work, or other demonstrated impact.
  • Ability to communicate complex research concepts clearly and collaborate across research, engineering, and robotics disciplines.
  • Demonstrated technical leadership and ability to influence research direction beyond individual projects.

Nice To Haves

  • Relevant experience in foundation models for physical AI.
  • Relevant experience in world models.
  • Relevant experience in multimodal foundation models.
  • Relevant experience in self-supervised and representation learning.
  • Relevant experience in video understanding and generative video models.
  • Relevant experience in spatial and temporal reasoning.
  • Relevant experience in predictive and generative modeling.
  • Relevant experience in embodied AI and robot learning.
  • Relevant experience in vision-language and vision-language-action models.
  • Relevant experience in reinforcement learning and learned decision-making.
  • Relevant experience in large-scale model pretraining.
  • Relevant experience in autonomous systems and physical-world intelligence.

Responsibilities

  • Conduct research and develop novel machine learning architectures, algorithms, and training methodologies for physical AI.
  • Advance foundational approaches to multimodal learning, representation learning, world models, and embodied intelligence.
  • Explore architectures for learning spatial, temporal, semantic, and physical representations from large-scale data.
  • Develop approaches for learning from video, images, sensor data, simulated environments, and other multimodal inputs.
  • Research self-supervised, generative, predictive, and other learning methods that improve models' understanding of the physical world.
  • Explore how foundation models can reason about dynamic environments, predict future states, and support downstream decision-making.
  • Investigate approaches connecting perception, learned world representations, planning, and action generation.
  • Develop model architectures and training strategies that improve generalization, robustness, scalability, and learning efficiency.
  • Design and execute rigorous experiments to evaluate research ideas and understand model behavior.
  • Develop evaluation methodologies and benchmarks for physical-world understanding, reasoning, prediction, and generalization.
  • Work with large-scale datasets and distributed training infrastructure to develop and evaluate models.
  • Translate promising research into prototypes and collaborate with Applied AI and engineering teams on real-world applications.
  • Stay informed about advances in foundation models, multimodal learning, robotics, and physical AI, and identify opportunities to advance Atoms' research direction.
  • Provide technical leadership through research direction, architecture reviews, mentorship, experimentation, and hands-on development.
  • Help establish the research culture, technical standards, and long-term direction of Atoms' growing AI organization.

Benefits

  • Medical insurance
  • Dental insurance
  • Vision insurance
  • Disability insurance
  • Life insurance
  • Flexible Spending Account and Health Savings Account options
  • 401(k)
  • Equity awards may be available
  • Sick time
  • Unlimited flexible time off
  • Paid holidays
  • Paid parental leave
  • Pre-tax commuter benefit plan
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service