VLA for Physical AI (Talent pool)

42dot is building the future of autonomous mobility, and we are growing our network of exceptional engineers and researchers for VLA for Physical AI roles based in San Francisco, California. This talent pool posting helps us connect with specialists in Vision-Language-Action modeling ahead of upcoming openings.

About the Role

As a VLA for Physical AI contributor at 42dot, you will help shape multimodal driving systems that link visual perception, scene understanding, and driving behavior into reliable autonomous action. This is a full-time, on-site position in San Francisco for people who want to push the boundaries of decision-making and action generation in physical environments.

Key Responsibilities

• Develop and refine Vision-Language-Action (VLA) style models that translate visual input and contextual cues into safe, robust driving decisions.
• Advance planning, decision-making, and action-generation capabilities for real-world driving scenarios.
• Integrate prior knowledge into model design to strengthen generalization across diverse conditions.
• Collaborate with perception, planning, and research teams to bring multimodal models from concept to deployment.
• Evaluate model performance and iterate on approaches that improve reliability in complex, dynamic settings.

Qualifications

• Strong background in machine learning, computer vision, or multimodal modeling, with an interest in Physical AI and autonomous systems.
• Familiarity with VLA-style architectures and the connection between perception, language, and action.
• Experience designing or training models that operate in real-world, embodied environments is a plus.
• Ability to work effectively within a cross-functional, research-driven team.
• Willingness to work full-time on-site in San Francisco, California.

About 42dot

42dot creates multimodal driving models that unite visual perception, scene context, and driving actions to deliver dependable autonomous behavior. Our mission is to advance how vehicles plan, decide, and act in real-world conditions by applying VLA-style modeling and leveraging prior knowledge for stronger generalization. We are assembling a team of the best minds in Physical AI to make that vision real.

How to Apply

If the VLA for Physical AI role at 42dot in San Francisco matches your expertise, we encourage you to join our talent pool. Submit your application through the official listing to be considered as opportunities open. View the full listing and apply.