Hacker News: bryantnychttps://news.ycombinator.com/user?id=bryantnycHacker News RSShttps://hnrss.org/hnrss v2.1.1Mon, 27 Apr 2026 11:49:01 +0000<![CDATA[New comment by bryantnyc in "IDEs we had 30 years ago and lost (2023)"]]>Anyone else on here recall IBM VisualAge for Smalltalk -> VAST, or Cincom Smalltalk?

]]>Sun, 19 Oct 2025 03:26:01 +0000https://news.ycombinator.com/item?id=45631896bryantnychttps://news.ycombinator.com/item?id=45631896https://news.ycombinator.com/item?id=45631896<![CDATA[New comment by bryantnyc in "Two things LLM coding agents are still bad at"]]>"...Recently, we did an experiment where we had a “red team” deliberately introduce an alignment issue into a model (say, a tendency for the model to exploit a loophole in a task..." https://www.darioamodei.com/post/the-urgency-of-interpretabi...

Anyone care to wager if anthropic is red teaming in production on paying users?

]]>Fri, 10 Oct 2025 00:47:22 +0000https://news.ycombinator.com/item?id=45534495bryantnychttps://news.ycombinator.com/item?id=45534495https://news.ycombinator.com/item?id=45534495<![CDATA[New comment by bryantnyc in "Two things LLM coding agents are still bad at"]]>"...very subtle and silently introduced mistakes are quite dangerous..."

In my view these perfectly serve the purpose of encouraging you to keep burning tokens for immediate revenue as well as potentially using you to train their next model at your expense.

]]>Fri, 10 Oct 2025 00:39:23 +0000https://news.ycombinator.com/item?id=45534453bryantnychttps://news.ycombinator.com/item?id=45534453https://news.ycombinator.com/item?id=45534453