Research Engineer, Pre-training

Lead and execute research-driven improvements to pretraining workflows for large language models, focusing on model architecture, data processing, and optimizer development.
Independently own smaller research projects while collaborating on larger initiatives.
Design, run, and analyze experiments to advance understanding of pretraining dynamics.
Scale and optimize training infrastructure to improve efficiency, reliability, and fault tolerance.
Develop and improve developer tooling to boost team productivity and enable end-to-end experimentation.
Contribute across the stack, from low-level optimizations to high-level model design, with emphasis on safety, steerability, and alignment with human values.