I'd further say that there are probably enough rational actors running evals out there that the marginally better is not pure vibes for the cases where people are spending lots of money, but I only have direct line of sight to some of those eval suites. Maybe everyone is irrational and anthropic is exploiting that!

(I should go measure this now, I'm curious)

For evals in particular (tuning workflows that agents are using), effectively not having to worry about price is an incredible multiplier - getting statistical significant signal is not cheap otherwise.

What people often want:

- https://trilogydata.dev/

I find those big stored procedures usually fall into two categories; logic that should be in the DB, but should be decomposed (staging tables, other SPs, etc) in which case they can be understandable in chunks; or logic that shouldn't be in the DB but has been shoved in there, in which case there's more of an ideological debate but I generally prefer to pull out and run in the application layer. (the latter is pretty much IMO the things that you've done after you've gotten the data at the right grain, when you are massaging it to a particular form/presentation format; performance is often the final arbitre here though).

Trilogy's model works quite well for agents, but I've avoided making AI features too native in the UI products - not everyone's cup of tea - so this month going to do a spike on a new pure AI native data authoring/exploration experience to see where that can go without messing with the core product, at both CLI/UI layers. Data consumption is (un?)fortunately a pretty perfect fit for the agent strengths with the right harness.

[1]https://trilogydata.dev/