Machine Learning Researcher

About Causal Labs

Causal Labs is building toward general causal intelligence — AI capable of predicting the future and identifying the optimal actions to change it. The bet is that achieving this requires a Large Physics Model (LPM), because domains governed by physics have inherent cause-and-effect relationships that visual and textual data alone cannot provide.

 Weather is the training ground. It is the most well-observed physical system on earth, offering rapid and objective ground truth feedback at a data scale that dwarfs what is used to train today's LLMs. The team believes this is the right wedge into a much larger problem.

 Causal Labs is built by researchers and engineers from Google DeepMind, Cruise, Waymo, Insitro, and Nabla Bio — people who have worked on self-driving, drug discovery, and robotics, and who believe general causal intelligence will be among the most important technical breakthroughs for civilization.

The Opportunity

Causal Labs is looking for a Machine Learning Researcher to work across the full ML stack — data, model, evaluation, and infrastructure — on frontier problems in large-scale physics modeling. You will implement novel architectures, build petabyte-scale data pipelines, and iterate rapidly on experiments grounded in verifiable, real-world feedback.

 This is research with real stakes and fast feedback loops. If you are excited by unsolved problems and want your work to be grounded in observable physical reality rather than benchmarks, this is that role. Experience in language models, computer vision, robotics, or biology is all relevant — the team values the ability to learn quickly in unfamiliar domains as much as depth in any one area.

What You'll Do

• •      Work across the full ML stack including data, model architecture, evaluation, and training infrastructure
• •      Implement novel model architectures and training algorithms
• •      Build and optimize data pipelines for massive, petabyte-scale, multimodal datasets
• •      Rapidly iterate on experiments and ablation studies
• •      Stay current on research and bring new ideas directly into the work

You Should Have

• •      Strong grasp of machine learning fundamentals with depth in at least one core domain such as computer vision, sensor fusion, language models, or physics-informed neural networks
• •      Experience training large-scale models and analyzing experiment results through careful ablation studies
• •      Experience writing and optimizing petabyte-scale data pipelines
• •      Familiarity with distributed training and inference

Nice to Have

• •      Familiarity with meteorology, computational fluid dynamics, or numerical simulations
• •      Experience training large models from scratch in frontier research settings