Staff Machine Learning Engineer - World Models City

Atoms•San Francisco, CA

2d•$273,000 - $321,000•Onsite

About The Position

Atoms is building Physical AI—real-world robots for industries like food, mining, and transport. Our systems are designed to understand, predict, and control physical operations, making them more reliable, scalable, and productive. This role focuses on developing world models and foundation models for physical AI, enabling machines to learn rich representations of the physical world from multimodal data. You will work at the intersection of foundation models, multimodal learning, computer vision, robotics, and embodied AI to define a new generation of physical AI systems. This is a deeply technical individual contributor role with significant influence over research direction and long-term AI architecture.

Requirements

Deep expertise in machine learning with experience developing large-scale deep learning or foundation model systems.
Strong understanding of modern model architectures and representation learning.
Experience with one or more areas such as multimodal learning, video models, generative models, self-supervised learning, predictive models, spatial intelligence, or embodied AI.
Experience training models on large-scale datasets and understanding the relationship between data, architecture, compute, and model performance.
Strong understanding of the full ML lifecycle, including data strategy, model architecture, training, evaluation, optimization, and inference.
Experience translating research ideas into functioning machine learning systems.
Strong software engineering fundamentals and the ability to remain deeply handson in Python and modern ML frameworks.
Demonstrated ability to operate in ambiguous research spaces where the architecture and solution may not yet be known.
A track record of making consequential technical decisions and influencing research or engineering direction beyond an individual project.
Ability to communicate complex research and technical ideas clearly and collaborate across research, engineering, and robotics disciplines.

Responsibilities

Define and help build Atoms' technical architecture for world models and foundation models for physical AI.
Develop large scale models that learn representations of complex, dynamic physical environments.
Build models capable of learning from multimodal inputs including video, images, spatial information, sensor data, robot state, and other realworld signals.
Explore architectures that capture spatial, temporal, semantic, and physical relationships within realworld environments.
Develop approaches for learning how environments evolve over time and how actions influence future states.
Research and build self supervised, generative, predictive, and representation earning approaches for physical world intelligence.
Explore the application of modern foundation model architectures to robotics and autonomous systems.
Develop training strategies that take advantage of largescale realworld and simulated datasets.
Make architectural decisions spanning data, model design, pretraining, finetuning, evaluation, inference, and deployment.
Establish evaluation methodologies for measuring a model's ability to understand, represent, and predict the physical world.
Partner closely with engineers and researchers working across perception, action models, robotics, autonomy, simulation, and ML infrastructure.
Translate emerging research into systems capable of operating on real machines in real environments.
Provide technical leadership through research direction, architecture reviews, mentorship, experimentation, and handson engineering.
Help establish the technical bar for the growing AI Research organization and participate in identifying and assessing exceptional engineering and research talent.