We're open sourcing Synthdocs. It's a synthetic training dataset of 50k synthetic documents from regulated industries we routinely work with.
yay ugly documents!
@Ai_YashAgrawal We think your agents should just be prompt files in your repo. We host, run, and track every one. You can start from plain text or 100+ templates.
@southpolesteve Agreed! We built our own deployment management system that allows any dev to spin up our entire constellation of workers, directly with Workers CI. it's such a killer DX move for the team.
Devs deploy their branch and can bind to remote from their dev server.
it took... 4hrs.
@lennysan This is great. We focused on signing enterprise tech clients from the outset for this reason. "CTO of x tech company chose our API because they couldn't develop the tooling internally" is a much stronger MOAT than "We blasted $100k in ad spend to get 1k signups"
Ultimate benchmark of the decision models: Jev (Typesafe) vs @millisecondsai.
Looks like Milliseconds always wins because we're so... fast... and make really good decisions?
I don't know tennis, so help me out here