Skip to content
SpinorML

About us

A software company in Birmingham

SpinorML builds teenygrad, a compiler that turns Rust into GPU kernels so a model ships to a device as one binary. The compiler is open source under Apache 2.0; the company sells commercial licences for the vision SDK and takes on consulting and training work.

Portrait of [TODO: FOUNDER NAME], founder of SpinorML

Who builds it

[TODO: FOUNDER NAME]

[TODO: FOUNDER NAME] has spent [TODO: N] years on the layer between models and silicon: CUDA kernel optimisation, MLIR compiler passes, Triton, and systems Rust. Before SpinorML, [TODO: ONE LINE ON PRIOR ROLES OR EMPLOYERS]. teenygrad comes out of deploying models to embedded and edge targets, where a Python runtime is not an option.

  • [TODO: COMMITTEE MEMBERSHIP — e.g. member of the … working group]
  • [TODO: CONFERENCE APPEARANCES — e.g. spoke at … 2025, … 2026]
  • [TODO: YEARS OF RELEVANT EXPERIENCE — e.g. 12 years in compilers and GPU work]

LinkedIn ↗

Licensing · Partnerships · Support

Contact us

A licence, a partnership, or a hard deployment problem — your message reaches someone who works on the compiler.

* required