ML Systems Engineer, Model Architectures (Mid/Senior/Principal)

Role overview Bring emerging machine-learning model architectures to an accelerator simulator before the underlying hardware is built. Working with chip architecture, compiler, SDK, and customer-facing teams, you will implement and verify the operations, numerical behavior, and memory layouts models require. The goal is accurate model execution in simulation and a faster, increasingly automated path from a new architecture to a working deployment.

Responsibilities - Implement simulator operators, numerical behavior, and memory layouts, matching reference implementations precisely. - Extend coverage across inference prefill and decode, embeddings, attention variants, mixture-of-experts routing, and quantized formats. - Maintain reference comparisons and regression checks as models, simulators, and hardware evolve. - Evaluate new architectures ahead of hardware availability, identifying compute, memory, and data-movement needs. - Work with customers and internal engineering teams to bring up models and improve the repeatability of that process.

Requirements - Experience implementing machine-learning systems, model operators, or inference software. - Strong understanding of transformer inference, including attention and related model operations. - Ability to reason about numerical accuracy, tensor layouts, quantization, and reference validation. - Software engineering experience building maintainable simulator or model-support code. - Ability to work across model, compiler, SDK, and chip design concerns.

Benefits and work setup The position is remote-first and includes a company-provided computer and home office setup. Benefits include paid medical, dental, and vision coverage for employees and dependents, paid holidays, flexible paid time off, equity, and a retirement plan with company matching.