We are building a system that represents domain knowledge as modular probabilistic models. Users compose them into larger structures, the system enforces consistency across them, and uncertainty is propagated across model boundaries autonomously by the composition itself.

You will implement models that our users need, inside of our framework. Our current applications range from equity valuation and financial distress monitoring to particle physics.

The interesting part is not any single model. It is that the same compositional machinery has to carry models from very different domains (finance, physics, others) without special-casing any of them. That generality is the hard part of the design, and it is what we are building toward.

What you will do

• Work with the product team to define what a model needs to do, then own it through to something a user can run

• Define and document the implementation approach for harder modelling problems

• Write code that is clean enough to be maintained by others and fast enough to run at the scale users need

• Establish that a model is correct, not just that it executes

• Find where the