Good catch! We're looking to add function calling support very soon, and have an open issue for it on our GitHub. If you want to raise a PR and add it, we'll help you land it and get it merged
This is really cool! When we were trying to launch the GSPMD feature for PyTorch/XLA at Google, one of our biggest bottlenecks was network overhead, but we didn't really have any robust tools to dig into it and perform root cause analysis. I'm loving the tools I see come out of Trainy.
We've actually been in contact with the qdrant team about adding it to our roadmap! Andre (CEO) was asking for an integration. If you want to work on the PR, we'd be happy to work with you and get that merged in
We offer auto-evals as one tool in the toolbox. We also consider structured output validations, semantic similarity to an expected result, and manual feedback gathering. If anything, I've seen that people are more skeptical of LLM auto-eval because of the inherent circularity, rather than over-trusting it.
Do you have any suggestions for other evaluation methods we should add? We just got started in July and we're eager to incorporate feedback and keep building.
For now, we just aggregate those across the models / prompts / templates you're evaluating so that you can get an aggregate score. You can export to CSV, JSON, MongoDB, or Markdown files, and we're working on more persistence features so that you can get a history of which models / prompts / templates you gave the best scores to, and keep track of your manual evaluations over time.