This sounds a lot like Spotify's XIRP. They managed to vibe something useful for their own engineers, were super hyped, and decided to release it in public. However what are the odds this harness will stand the test of time against actual harness makers like OpenAI or Anthropic? Neither Spotify or IBM are focused on making models or harnesses, these are just side-products.
> actual harness makers I agree IBM Bob is not the revolution, but… We are still in the punch card era of LLMs. Maybe even the Altair era. None of this stuff has really been figured out yet.
IBM Bob
61–70 of 348 posts
Re: IBM Bob
#62Should have been IBM Tom, after legendary IBM CEO Thomas J. Watson. But perhaps they didn't want to sully his name.
Re: IBM Bob
#63Re: IBM Bob
#64Re: IBM Bob
#65I wish IBM got better at product again, and took themselves off the services and consulting morphine they've been on for ~35 years now since OS/2 Warp flopped. Their 1980s/1990s hardware was great. Their software actually was great until Win95 took their lunch money. They've just slunk off into a corner hugging their old architectures they have a moat around, and thrown out some audacious nonsense like Watson now and again, but nothing has stuck. When is somebody in that organisation going to try and innovate properly again?
I'm not sure this is the answer I'd be rooting for.
Re: IBM Bob
#66Looks awful from a design perspective. Had to double check that this was the actual IBM domain
Re: IBM Bob
#67Re: IBM Bob
#68Re: IBM Bob
#69The "What developers say" carousel has a single Java dev and dev advocate. They got lots of quotes from managers, directors, CEOs, and an "AI strategist".