About this role

Our mission is general causal intelligence; AI that is capable of (1) predicting the future and (2) identifying the actions to alter it.

To achieve this breakthrough, we are building a Large Physics foundation Model (LPM) because physical systems, unlike text or images, are governed by verifiable cause and effect. We believe that scaling on physics will enable an understanding of causality required to predict and control physical systems, starting with weather.

Our founding team has built and deployed AI against the physical world in robotics, drug discovery, and particle physics at institutions like DeepMind, Waymo, Cruise, Insitro, Nabla Bio, and CERN.

We look for research engineers who are excited to tackle unsolved problems. Progress is only as trustworthy as its measurement. As our model, data, and reasoning efforts multiply, every team needs to know, precisely and comparably, whether a change made the model better. Your mission is to build the central evaluation framework that the entire research organization runs on: the pipelines, the metrics, and the tools that turn results into shared understanding.

Responsibilities

Design and build a central, reusable evaluation framework that every model and every team runs through
Implement evaluation pipelines, benchmark suites, and baselines that make model quality measurable and comparable across efforts
Build the visualization and dashboard tools that turn raw results into shared, actionable understanding for the whole team
Establish sound statistical methodology for evaluation, so teams can distinguish real improvements from noise
Partner with research and domain teams to translate what "good" means in each domain into standardized, automated metrics
What we're looking for
We value a relentless approach to problem-solving, rapid execution, and the ability to quickly learn in unfamiliar domains.
Strong software engineering skills and experience building data or evaluation pipelines at scale
Experience turning research or model outputs into metrics, benchmarks, and visualizations that teams rely on
Solid grasp of probability and statistics, with the judgment to design evaluations that measure what they claim to
Full-stack range: comfortable building both backend pipelines and the frontend tools people read results in
Owns deliverables end-to-end, from collecting requirements to autonomously driving execution