Founding Machine Learning - World Models

One Robot (YC W26); San Francisco (CA); On-site; USD 150,000 - 275,000; Full time

Job summary

One Robot (YC W26) seeks an engineer to build world models that simulate manipulation scenes and validate policies in a scalable way. You will develop generative models with controllability and physical fidelity matching real-robot behavior.

You will train video and dynamics models on multi-GPU clusters, own data pipelines, and ensure learning across customers and tasks. Strong Python/PyTorch skills and video generation and 3D vision experience are required.

Qualifications

• Strong coding in Python and PyTorch.
• Video generation experience end-to-end.
• Track record of large-scale training on cluster systems.
• Working knowledge of 3D vision, multi-view geometry, scene reconstruction.

Responsibilities

• Train video and dynamics models with action conditioning for manipulation policies.
• Push long-horizon coherence by developing architectures and training methods for hard tasks.
• Own training infrastructure: run multi-GPU clusters, write custom CUDA, debug at scale.
• Build a world-model data engine to enable learning across customers and manipulation tasks.

Skills

Python

PyTorch

Video generation experience

Large-scale training

3D vision

Tools

CUDA

Job description

We build world models that simulate manipulation scenes faithfully enough to validate, and one day, train policies without touching a robot. You'll develop generative models that make this work, with the controllability and physical fidelity to match real-robot behavior.

What you'll do:

• Train video and dynamics models: Develop world models with action conditioning for manipulation policies.
• Push long-horizon coherence: Develop architectures and training methods that extend rollout quality on hard physical tasks.
• Own training infrastructure: Run multi-GPU clusters, write custom CUDA, debug at scale.
• Build the world-model data engine: Design, implement, and improve a data engine that allows the world model to compound learning across customers and manipulation tasks.

Requirements:

• Very strong coding in Python and PyTorch (or similar).
• Video generation experience: Deep experience training image or video generation models end-to-end.
• Large-scale training: Track record operating training runs at cluster scale.
• 3D vision: Working knowledge of multi-view geometry, scene reconstruction, and physical priors.

Compensation Range: $150K - $275K