Distributed Training Engineer for Transformers

Axiōma Search in London is recruiting an IC to own and scale distributed training infrastructure for large transformer models. You will push capabilities from tens to hundreds of GPUs, drive efficiency, and shape the data pipelines, networking, and storage to support ever-larger runs.

This hands-on role demands independent coding and the ability to lead infrastructure decisions as the product evolves. You will work on-site five days a week in London, with visa sponsorship and relocation options