Re:How the Claude Code team thinks about verification
by jeffybefffy519 on Tuesday November 17, 2026 @05:08PM
(#50538938)
I feel the ironic thing about this end-to-end development style with no human in the loop method that a lot of shops are going for is that it really places a lot more emphasis on manual user acceptance testing.
Otherwise the type of engagement loop that you get from just clauding every feature end to end pushes you to drop important quality steps because the agent is covering it but often it just isn't and you get slop with regressions everywhere.
TLDR we think of the frontier as "the best"... but really its the most unstable in this new world - Claude Code, Codex and their supporting models absolutely show this time and time again.
Otherwise the type of engagement loop that you get from just clauding every feature end to end pushes you to drop important quality steps because the agent is covering it but often it just isn't and you get slop with regressions everywhere.
TLDR we think of the frontier as "the best"... but really its the most unstable in this new world - Claude Code, Codex and their supporting models absolutely show this time and time again.