Robotics Researcher: 3D Vision & World-Action Models

Digitailor; San Francisco (CA); On-site; USD 150,000 - 210,000; Full time

Benefits offered by this job

401(k) and full health insurance

Meals provided (breakfast, lunch and 3

Ubers covered home

All roles onsite in SF HQ

Job summary

Sieve seeks a Robotics Researcher to lead training and evaluation of world-action models and 3D perception systems for robot learning. You will connect data capture, curation, and labeling to measurable improvements in robot capabilities, running projects from question to answer.

Expect to work with 3D data, simulation tools, and real-world testing, collaborating with engineers and researchers at partner labs. This role is based in San Francisco and requires onsite presence.

Qualifications

• 4+ years of research or engineering experience in robot learning, embodied AI, or 3D computer vision.
• PhD welcomed but not required; strong publications or open-source projects count.
• Fluent Python plus PyTorch or JAX: you can write and modify training code.
• Hands-on 3D vision work: multi-view geometry, depth, pose, tracking, or reconstruction.
• Rigorous on experimentation: baselines, ablations, and clear conclusions.

Responsibilities

• Data ablations for robot learning and evaluating how data sources affect policies and world models.
• Train and post-train world models and policies to follow commanded actions and manipulate objects.
• Build pipelines for 3D labeling: depth, camera trajectories, pose, and scene reconstruction.
• Design and run evaluation protocols with fixed test tasks and environments.
• Publish findings through papers, technical reports, blogs, and benchmarks.

Skills

Python

PyTorch

JAX

3D computer vision

Experiment design

communication skills

Education

PhD welcomed

Tools

MuJoCo

Isaac Sim

Job description

Sieve seeks a Robotics Researcher to lead training and evaluation of world-action models and 3D perception systems for robot learning. You will connect data capture, curation, and labeling to measurable improvements in robot capabilities, running projects from question to answer.

Expect to work with 3D data, simulation tools, and real-world testing, collaborating with engineers and researchers at partner labs. This role is based in San Francisco and requires onsite presence.