Alexander L.

FYI I decided to go with Hugging Faces Spaces as a managed service for hosting and GPU support, seems like that will be easier than Kubernetes. Hopefully I'll be sharing the link to the released app in the 🔒[private feed] by next week 🤞

Thanks Jason, Tammy L. and Aparna D.! I need to get through a cloud deployment process before I can bring put my app into production and get that sweet wine of Arize observability diagnostics. 🤓 I made a post here with more details: https://arize-ai.slack.com/archives/C017C8Z8GBC/p1716567646862959 if you know anyone who's good with Cloud please send them my way! Thank you 🙇

Welcome Francisco! How's the AI scene in Japan? I'm following Sakana.AI closely

Sorry for the delayed reply -- I can do that! I'm in Mountain time so I'll block my calendar at 2pm.

Thanks for quick response, and yes would be good to connect. Tuesday's better on my end, any particular time preferred for you?

Thank you! 🤓

Alternately, should I be structuring my data a different way to align with the built in dashboard use cases like time series? (maybe setting up a Monitor is a better way to go)?

My question is, how can i use Arize dashboard to visualize this data? Specifically I want to display the prediction_id and other metadata (city, country, prompt, etc) for rows that show a content violation. I'm having trouble setting that up in the dashboards