Physical AI Research Engineer

Redapto; San Francisco (CA); Hybrid; USD 130,000 - 190,000; Full time

Job summary

Redapto in San Francisco is seeking a research-focused professional to advance how multimodal agents understand, simulate, and act in dynamic physical environments. You will explore multimodal reasoning, world models, planning, and closed-loop evaluation to create robust prototypes and production capabilities.

Collaborate with platform and deployment teams to validate research in the field, design meaningful benchmarks, and contribute to the growth of a high-ownership, high-autonomy team that

Qualifications

• Strong analytical foundations developed in high-performing, fast-paced environments.
• Ability to handle ambiguity, form clear mental models quickly, and own problems.
• Energised by ambitious, precise, owner-driven culture and accountability.
• Clear, concise communication, both written and spoken.

Responsibilities

• Research multimodal reasoning, world models, planning, and evaluation.
• Turn promising methods into robust prototypes and production capabilities.
• Design benchmarks grounded in real operational decisions and failure modes.
• Collaborate with platform and deployment teams to validate research in the field.

Skills

Analytical Foundations

Action Bias

Culture & Velocity Fit

Communication & Reasoning Clarity

Job description

Redapto is building the agentic operating platform for the physical world. We help teams in manufacturing, logistics, energy, aerospace, and defense turn fragmented systems and operational knowledge into governed AI workflows that act in production.

Our work sits where software meets real operations: machines, fleets, suppliers, field teams, safety constraints, and decisions with consequences. We build for dependable outcomes, not demos.

We're a small, high-ownership team spanning Bangalore and San Francisco. We value first-principles thinking, technical depth, direct customer contact, and the craft required to make frontier AI trustworthy in the real world.

About the Role

Advance how multimodal agents understand, simulate, and act in dynamic physical environments.

Responsibilities

• Research multimodal reasoning, world models, planning, and closed-loop agent evaluation.
• Turn promising methods into robust prototypes and production capabilities.
• Design benchmarks grounded in real operational decisions and failure modes.
• Collaborate with platform and deployment teams to validate research in the field.

What We're Looking For

Analytical FoundationsA strong analytical foundation built through high-performing environments that demand rigorous thinking, structured problem-solving, and rapid synthesis of information.

Action BiasYou thrive in ambiguity, form clear mental models quickly, take ownership instinctively, and enjoy diving deep into unfamiliar software categories.

Culture & Velocity FitYou're energised by environments defined by ambition, precision, ownership, integrity, and responsibility.

Communication & Reasoning ClarityYou communicate in a structured, concise, and compelling way — verbally and in writing.

Why Join Redapto

• Build foundational technology for industries with enormous real-world impact.
• Work in a high-trust, high-autonomy environment with direct ownership.
• Learn across frontier AI, enterprise systems, and complex physical operations.

Our Process

Joining Redapto means stepping into a culture that values excellence, curiosity, and potential above all else. Our selection process reflects that.

If you excel at our initial screening, you move forward to interviews that dive deeper into your ambitions, mindset, and how you can shape growth with us. You'll take on a few challenges designed to see how you think, learn, and adapt under pressure.

We set the bar high and only make offers when we're confident it's an extraordinary fit—for both you and Redapto.

Redapto is proud to be an equal opportunity employer. We celebrate diversity, foster inclusion, and ensure every candidate has equitable access to succeed.