Elena Samuylova

We use HuggingFace models to allow evaluation of the quality of LLM outputs with local models and classifiers.

We host and develop our open-source tooling on GitHub.

We use OpenAI API as one of the options for implementing LLM as a judge evaluation functionality.