News for hackers. Stuff that matters.
Re:Claude Code is unusable for complex engineering tasks wit... by JamesSwift on Monday April 06, 2026 @03:49PM (#47666007)
I commented on the GH issue, but Ive had effort set to 'high' for however long its been available and had a marked decline since... checks notes... about 23 March according to slack messages I sent to the team to see if I was alone (I wasnt).

EDIT: actually the first glaring issue I remember was on 20 March where it hallucinated a full sha from a short sha while updating my github actions version pinning. That follows a pattern of it making really egregious assumptions about things without first validating or checking. Ive also had it answer with hallucinated information instead of looking online first (to a higher degree than Ive been used to after using these models daily for the past ~6 months)

Re:Claude Code is unusable for complex engineering tasks wit... by dev_l1x_be on Monday April 06, 2026 @04:25PM (#47666560)
It hallucinated a GUID for me instead of using the one in the RFC for webscokets. Fun part was that the beginning was the same. Then it hardcoded the unit tests to be green with the wrong GUID.
Re:Claude Code is unusable for complex engineering tasks wit... by viktorianer on Tuesday April 07, 2026 @09:36AM (#47675184)
The hallucinated GUIDs are a class of failure that prompt instructions will never reliably prevent. The fix that worked: regex patterns running on every file the agent produces, before anything executes.
Re:Claude Code is unusable for complex engineering tasks wit... by JamesSwift on Tuesday April 07, 2026 @10:06AM (#47675611)
Well Ive never had the issue before and have hit that / similar issues every few days over the past couple weeks.