DHolzer

> Totally disagree here, using embeddings is much more reliable / robust, I wouldn't put much stock in LLM output, too much going on

I think both ways can be the preferable option, depending on how well the embedding space represents the text - and that is mostly dependet on the specific use case and model combination.

So if the embedding space does not correctly project required nuance, then it's often a viable option to get the top_n results and do the rest by utilizing the llm + validation calls.

But i do agree with you, i would always like to work with embeddings rather than some llm output. I think it would be such a great thing to have rock solid embedding space where one would not even consider to look at token predictor models.

I work in AI and love the technology. But all the hype and grandiose claims make it awkward when people ask what I do, and it makes hiring harder when experienced developers hear 'AI development' and walk away - even though it's mostly just solid full-stack engineering work.

Any negative aspect of media from the past can, and often will, be transformed into a positive trait in future media.

People embrace vinyl records in an age of digital music. They take photos with analog cameras even though everyone has a phone in their pocket. Musicians use the harsh artifacts of MP3 compression as creative effects in their music. The examples are countless, and they all emerge precisely when the media that once produced these unwanted artifacts becomes obsolete.

If you haven't noticed this shift, I suggest you learn to recognize it quickly. Otherwise, you might miss out on great content because it doesn't make it past your mental spam filter.

And if you don't want to adapt, that's fine too—just don't tell others how to manage their websites.

When I read "open source" i thought they actually are doing open source instead of "open weights" this time. Surely they would expect to be called out on hackernews if they label it incorrectly...

Thanks for pointing that out @Hizonener

I share your perspective on pricing. I had a discussion with my team lead about why we haven't taken on the task of running our own machines. The rationale behind it is that while server management may seem easy, ensuring its security can be complex. Even in a sizable company, it's worth considering whether you want to shoulder the responsibility or simply pay a higher cost to have Microsoft handle it. Personally, I love hosting my own infrastructure. It's fun, potentially saves me some cash, allows me to learn a lot, and gives me full control. However, I understand now that on a company scale, others may see it differently.

--edit--

I forgot to add the following: that's 32k if you run the system 24/7. Usually it's up for a few hours per month, so you end up paying maybe 2k for the whole year.

in Azure that's roughly 5k per year if you pay for the whole year upfront. I have the pleasure of playing with 64cores, 256gb RAM and 2xV100 for data science projects every now and then. That turns out to be roughly 32k per year.

I love phones with keyboards, I've been using a terminal style launcher (https://github.com/fAndreuzzi/TUI-ConsoleLauncher) for almost 6 years now, but I really can't get behind the Unihertz devices. The components and the build quality are just so weak compared to the Blackberry Key 1 and 2.. That trend seems to hold true for the non-keyboard Unihertz devices too. Sadly, the Key1&2 are a little slow and don't get updates, so i am back using touch screens again.