nathan gonzalez

Any ideas around token usage for this? I noticed you're using the anthropic computer usage tooling, which is really neat, I wonder how much variance in each test run there is. I imagine there could be a use case for caching the actual test tool utilizations and rerunning them. seems like having the tests be more deterministic (which i assume would be a natural result of that) would be pretty...

I like this idea, but I hate the onboarding order. the fact that I go through the entire onboarding process before seeing a pricetag feels a little predatory, now I have an app with access to all my screentime information that I am not sure if I want to pay for but I've already given all the permissions to.

Does this handle anything with rate limiting, or does it allow for multiple instances of the same kind of model? we have azure openai deployments in multiple regions for instance.

What is the benefit to using this vs using advanced analysis in chatgpt?

Is this using LangChain under the covers? The format of the responses makes me feel like it is.

Any support for azure?

interesting idea but a vast majority of the domains seem to be sexual or pejorative in nature.

?makers Knowing how a lot of the larger enterprise systems like ADP, Paycor and Paycom work from integrating with them in the past I'm curious how you could possibly integrate with them in this manner. If you've got that part figured out, kudos, I'm sure it was difficult and I don't envy the review processes you likely went through.

sooo expensive. I can't imagine them being worth it.

?makers is this data available for consumption anywhere? are you validating against the cdc or jhu datasets at all?

Not gonna lie, commenting for the discount.

any plans on supporting sql server?