News for hackers. Stuff that matters.
Re:Anthropic hits $2T in Nasdaq debut by lbrito on Thursday November 19, 2026 @01:34PM (#50564154)
>gas town failed quietly

This is a big problem. When this kind of thing gets launched, its all fanfare, this is the future, get on the boat or die.

But then it fails miserably and its like it never existed

Re:Anthropic hits $2T in Nasdaq debut by skydhash on Thursday November 19, 2026 @01:42PM (#50564295)
Remember prompt engineering? Even skills and specs seems to be slowly dying.
Re:Anthropic hits $2T in Nasdaq debut by oblio on Thursday November 19, 2026 @02:07PM (#50564693)
What's happening to skills?
Re:Anthropic hits $2T in Nasdaq debut by jmartrican on Thursday November 19, 2026 @02:18PM (#50564858)
I see skills as bash scripts. they are useful, even small ones.
Re:Anthropic hits $2T in Nasdaq debut by _aavaa_ on Thursday November 19, 2026 @02:23PM (#50564935)
To the extent that they are it's because they are no longer needed since the models do more of those things on their own without being prompted to do so.
Re:Anthropic hits $2T in Nasdaq debut by 20k on Thursday November 19, 2026 @02:32PM (#50565047)
There was never any evidence that they did anything all that useful
Re:Anthropic hits $2T in Nasdaq debut by _aavaa_ on Thursday November 19, 2026 @02:41PM (#50565187)
There was never any evidence that writing a spec and having the model built against that spec was useful??
Re:Anthropic hits $2T in Nasdaq debut by Rapzid on Thursday November 19, 2026 @04:06PM (#50566361)
Prompt engineering as the sole means of getting anything of value and quality is a thing of the past.

But I also see people complaining about frontier models writing too many helper functions, changing more code than necessary, and etc.

These are things that can be addressed through context engineering; AGENTS.md, skills, etc.

Also a lot of use cases require smaller, lower latency models. And these faster, smaller models to this day exhibit similar failure modes to 2-3 years ago and require very purposeful context engineering to get satisfactory results.