Research Evaluation Lead

Socket.dev; Mountain View (CA); On-site; USD 120,000 - 180,000; Full time

Job summary

Rhoda AI is seeking a researcher to own end-to-end eval workflows for cutting-edge robot models. You will translate research questions into clear protocols, run pilots, and drive QA, ensuring consistent results across shifts and stations.

The role emphasizes hands-on coding, rigorous experimentation, and the ability to distinguish model failures from hardware, data, or setup issues. Collaboration with multiple teams is essential.

Qualifications

• Familiarity with robotics or ML experimentation.
• Background in computer science or hands-on coding experience.
• Detail-oriented and able to interpret experimental intent.
• Demonstrated ownership and execution capability.
• Experience with robotics testing, data collection, lab operations, or QA preferred.

Responsibilities

• Translate research intent into eval protocols and success criteria.
• Own end-to-end evaluation workflow from model handoff to QA and results.
• Train and manage eval pilots across shifts and stations.
• Distinguish model failures from hardware, setup, or data issues.
• Maintain eval setups, resets, randomization, and experiment traceability.
• Track quality, throughput, and bottlenecks; improve the eval process.
• Collaborate with Research, Robot Data, and Eval Platform teams.

Skills

Robotics / ML experimentation

Coding

Detail-oriented

Ownership

Robotics testing / QA

Education

Computer science background

Job description

At Rhoda AI, we’re building the next generation of generalist intelligent robots. We own the full robotics stack from high-performance hardware and robot systems to the infrastructure and state-of-the-art foundation world models that control our robots. Our robots are designed to be generalists capable of operating in complex, real-world environments and handling long-tail edge cases, made possible by our cutting edge research and end-to-end system design. We've raised over $450M and are investing aggressively in model research, infrastructure, hardware development, and manufacturing scale-up to make generalist robotics a reality.

Mission

Own end-to-end robot model evaluation for Research. Turn research questions into consistent, high-quality, repeatable evals that enable fast iteration and trusted results.

Responsibilities

• Translate research intent into clear eval protocols, trial plans, and success criteria.
• Own execution end-to-end: model handoff → station readiness → pilot execution → QA → results.
• Train and manage eval pilots; ensure consistency across people, shifts, and stations.
• Distinguish model failures from hardware, setup, operator, or data-quality issues.
• Maintain eval setups, resets, randomization, metadata, and experiment traceability.
• Track quality, throughput, and bottlenecks; continuously improve the eval process.
• Partner closely with Research, Robot Data, and Eval Platform teams.

What we’re looking for

• Some understanding of robotics / ML experimentation.
• Computer science background or hands-on experience with coding
• Rigorous, detail-oriented, and able to understand the intent behind an experiment, not just execute instructions.
• Strong hands-on execution and ownership.
• Experience in robotics testing, data collection, lab operations, or QA preferred.

Success looks like

A researcher can hand off a model and research question and receive a trusted, standardized eval result with sufficient trials and QA.