jobs

Data Scientist, Model Evaluation

Cohere is seeking a Data Scientist to develop and run evaluations for our language models. You will design benchmarks, build automated testing pipelines, and analyze model behavior to identify regressions and improvements.

Company

Cohere

Location

San Francisco, CA

Status

Closed

Posted

2026-02-15T00:00:00+00:00

Cohere is hiring a data scientist to work on evaluations for its language models. The job is to design benchmarks, build pipelines that run tests automatically, and study model behaviour to spot regressions and improvements. The posting names no team, level or visa support. It is listed as full-time, remote, based in San Francisco, with pay between 120,000 and 175,000 USD. Distinctive: a data science role centred entirely on measuring model quality rather than building models.