Member of Technical Staff, World Models

Baseline Robotics; Zürich; Vor Ort; CHF 120.000 - 160.000; Vollzeit

Benefits dieser Stelle

Stock options

Zusammenfassung

Baseline Robotics in Zurich is hiring to build action-conditioned world models from camera and sensor data, enabling planning with model rollouts and synthetic trials to broaden data coverage. You will diagnose divergences and trace causes to improve predictions, while contributing to a cutting-edge open-source robotics platform.

Ideal candidates have hands-on experience with world-models, model-based RL, Python and PyTorch, and familiarity with JAX or Rust, plus GPU-accelerated simulation

Qualifikationen

• Experience training world models, video prediction, or latent dynamics models on real data.
• Knowledge of model-based reinforcement learning, planning, and sequence modelling.
• Publications in top machine learning or robotics venues, or equivalent experience.
• Fluency in Python and PyTorch.
• Experience with JAX or Rust.
• Experience with GPU-accelerated simulation tools (Isaac Lab, Newton, mjlab, Genesis).

Aufgaben

• Build action-conditioned world models predicting dynamics from camera and sensor data.
• Use model rollouts for planning and post-training reference policies.
• Generate synthetic trials from the model to expand coverage where real data is scarce.
• Diagnose where model predictions diverge from the stations and trace the cause.

Kenntnisse

World models

Video prediction

Latent dynamics

Model-based RL

Planning

Sequence modelling

Publications

Python

PyTorch

JAX

Rust

GPU simulation

Tools

PyTorch

JAX

Rust

Isaac Lab

Newton

mjlab

Genesis

Jobbeschreibung

About Baseline

Open source revolutionised how we write software, but its impact on robotics stays limited due to expensive hardware accessible only to well-funded labs. We run that hardware as shared infrastructure for the broader community, billed only for active compute. For a given task, policies are evaluated under similar conditions, giving researchers common ground for fair comparisons.

In this role you will

• Build action-conditioned world models predicting dynamics from camera and sensor data.
• Use model rollouts for planning and post-training reference policies.
• Generate synthetic trials from the model to expand coverage where real data is scarce.
• Diagnose where model predictions diverge from the stations and trace the cause.

What we hope you'll bring

• Experience training world models, video prediction, or latent dynamics models on real data.
• Knowledge of model-based reinforcement learning, planning, and sequence modelling.
• Publications in top machine learning or robotics venues, or equivalent experience.
• Fluency in Python and PyTorch.
• Experience with JAX or Rust.
• Experience with GPU-accelerated simulation tools (Isaac Lab, Newton, mjlab, Genesis).

We encourage you to apply even if you do not believe you meet every qualification.

Details

• Location: Zurich, Switzerland, on-site 5 days a week.
• Annual salary: CHF 120,000 - 160,000.
• Every full-time hire receives stock options as part of the stated salary.

Deadline to apply: None. Applications will be reviewed on a rolling basis.