gradiatedCareers
Build inference from model to silicon.
Our goal is to maximize tokens per joule. We build software and hardware for open-weight model inference. We want people who can go deep across the abstraction stack.
How we work
Depth across the stack.
We work from first principles. We start by thinking about the model as a mathematical function. We map that function onto hardware. We then build a custom software stack for each model. This maximises performance and removes wasted abstractions.
We pay at the top of the market because talent density is what will allow us to build the most efficient inference. A small team of exceptional people can move faster, make better decisions, and solve harder problems.
Open roles
We are hiring.
Founding Inference Research Engineer
Build and optimize the complete inference stack for open-weight models on existing hardware.
Head of Silicon
Define our silicon direction and build a hands-on team for model-specific inference hardware.
Founding Silicon Engineer
Design model-specific inference hardware across architecture, implementation, and verification.