Elena Samuylova

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

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

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