Senior Research Scientist (Physical AI/Interpretability)

World Mechanics; San Francisco (CA); On-site; USD 300,000 - 500,000; Full time

Benefits offered by this job

Flexible working hours

Conference travel support

Office space and in-person collab

High ownership role

Real-world impact research

Job summary

World Mechanics is seeking a Senior Research Scientist (US) to lead end-to-end research programs across white-box evaluation, interpretable model training, and simulation for physical systems. You will set direction for multimodal world models, establish milestones, and mentor researchers while collaborating with leadership and customers to translate real needs into capabilities.

The role emphasizes empirical rigor, ownership, and high initiative in an early-stage company.

Qualifications

• Track record of high-impact ML research
• Ability to generate original research directions
• Startup alignment: high agency and fast iteration
• Pragmatic, empirical mindset with focus on real-world impact
• Strong writing and communication skills

Responsibilities

• Own end-to-end research programs aligned with core agenda from problem selection to validation and documentation
• Set technical direction across world models trained on video, sensor, time-series, and multimodal data
• Set milestones, de-risk unknowns early, and make high-quality calls on next steps
• Drive research practice: experimental design, iterations, ablations, evaluation, and write-ups
• Mentor researchers and engineers; provide technical guidance and potential management
• Partner with leadership/customers to translate needs into repeatable capabilities
• Lead and supervise publications and model releases

Job description

About World Mechanics

World Mechanics is a commercial R&D neo-lab developing interpretable foundation models of and for the physical world. We study what mechanisms world models learn about the systems they are trained on, and how to make those mechanisms more faithful to the underlying causal structure of those systems. We believe that models which learn interpretable, causal abstractions can provide a more principled foundation for explaining predictions, verifying model behavior, and monitoring and controlling models in deployment. Our goal is to build general-purpose world models that serve as reliable simulators of physical systems and can be deployed broadly across applications ranging from prediction and maintenance to robotics and scientific discovery.

We focus on three connected research directions:

1. white-box evaluation of physical models,




2. training intrinsically interpretable foundation models for the physical world,




3. interpretability-based learned simulators for physical systems.

What you’ll do

• Own end-to-end research programs aligned with our core agenda (white-box evaluation, intrinsically interpretable model training, and interpretability-based simulation), from problem selection to rigorous experimental validation to internal documentation and delivery.




• Set technical direction across world models trained on video, sensor, time-series, and other multimodal data, identifying which modeling and interpretability approaches generalize across physical domains.




• Set clear milestones, de-risk the hard unknowns early, and make high-quality calls on what to pursue next based on evidence.




• Drive strong research practice across the team: experimental design, iteration loops, ablations, evaluation methodology, and clear write-ups.




• Mentor and support other researchers and research engineers; provide technical guidance, feedback, and (as the team grows) direct management.




• Partner with leadership and (when relevant) customers to translate real needs into repeatable technical capabilities.




• Lead and supervise publications and model releases, including open-source releases, when aligned with our commercial and customer goals.

What we’re looking for

• Track record of high-impact ML research with strong taste and consistent execution.




• Ability to generate original research directions and deliver concrete results under resource constraints.




• Startup alignment: high agency, fast iteration, low ego, and comfortable wearing multiple hats.




• Pragmatic, empirical mindset; avoids safety/capability ideology and lab politics while still taking real-world risks seriously.




• Strong writing and communication; you can make research legible to technical and non-technical stakeholders.

Nice to have

• Prior work directly tied to mechanistic interpretability, multimodal/video world models, robotics foundation models, or scientific/industrial time-series.




• Experience translating research into product capabilities, or building platforms used by other researchers.

Our values

Empirical Rigor and Intellectual Openness

We approach research with an open mind and let evidence guide our decisions. We test assumptions, update our views when the facts change, and focus our efforts where they can have the greatest impact. We value clear thinking and practical progress over allegiance to particular schools of thought or debates that are not supported by meaningful evidence.

Ownership and Initiative

We seek people who are energized by the pace, autonomy, and ambiguity of an early-stage company. Our team members take responsibility beyond narrow job descriptions, move quickly from ideas to execution, and reliably carry work through to completion. We value genuine commitment to the mission and the desire to build, not simply to participate.

Seriousness, Kindness, and Mutual Respect

We aim to build a team of thoughtful, optimistic, and grounded people who care deeply about both the quality of their work and how they treat others. We keep ideological and political debates—including factional debates around AI safety and capabilities—outside the workplace and maintain our focus on the work itself. We welcome substantive disagreement while holding a high bar for respect, integrity, and professional conduct.

Research with Real-World Impact

We pursue research with the intention of translating it into useful products and capabilities. We value researchers who are enthusiastic about commercialization and want to see their work succeed beyond the research environment. Staying close to product, engineering, and - when useful - customers and real-world settings helps ensure that our research remains grounded in genuine needs and constraints.

Compensation

Senior Research Scientist (US):

• Base salary: $300,000 - $500,000




• Equity: Competitive; disclosed during process.

What we offer

• Flexible working hours




• Conference travel support, with opportunities to publish and open-source research when aligned with our commercial and customer goals




• Office space and in-person collaboration




• High-ownership role with substantial responsibility and autonomy




• Problems worth working on: research that has direct commercial and real-world safety/economic relevance