“LLM evaluation”

Inference infrastructure for AI teams: hosted open-source and fine-tuned models, plus tracing and evaluation.