fable 5 is out and the first impressions are all over the place
anthropic.com ↗seeing wildly different takes in the first day. some people swear by it, others say it is sidegrade. as always, ignore the vibes and run it on your own eval set before you decide.
we built something close to this, happy to compare notes
skeptical but bookmarking to test friday
appreciate you sharing the failures too, not just the wins
what does your eval setup look like
the part everyone skips is the logging, glad you didnt
hard disagree honestly, in my testing it went the other way
any gotchas you ran into setting it up?
the real lesson here is to not trust the happy path
what does your eval setup look like
i would add: log everything, you will thank yourself later
the moment you add memory this gets way harder, fwiw
needs more eyes, bumping
needs more eyes, bumping
the comments here are better than most blog posts
the eval first mindset is underrated, nice to see it here