They are off the charts. We're really proud of our baby.
The short answer is that we built a database from the ground up (only one external dependency) in Rust to leverage a universalized data architecture model we created. That's all we feel comfortable saying at the moment.
But if you're interested we can give you an API key to test it yourself (on any dataset you'd like) early next week. We're also going to drop a video on query features later this week, which I'll post here as well.
We also applied to YC, so any love here would be appreciated.
