thatcherthorn

As a backend developer who previously used bootstrap, it has been a delight to discover tailwind: https://tailwindcss.com/

They've reported surpassing GPT4 on several benchmarks. Does anyone know of these are hand picked examples or is this the new SOTA?

Comprehensive. Seems surprising that Harvard is sitting at the absolute bottom of the list.

As now f-strings can contain any valid Python expression inside expression components, it is now possible to nest f-strings arbitrarily:

>> f"{f"{f"{f"{f"{f"{1+1}"}"}"}"}"}"

'2'

Is anyone aware of the change to the interpreter that allows for this?

For what reasons?

I just discovered tailwind after using bootstrap for a few years...

Yes. It is amazing. Flowbite is the component library I've been using.

Interested in this clever tactic you speak of

This is a fantastic library. I plan to use some of the search functionality with a system that tries to figure out how to manipulate/work with/add features to existing code.

Curious, did you try GitHub copilot before switching to Amazon?

If you have, what are the differences?

AlloyDB AI allows users to easily transform their data into vector embeddings with a simple SQL function for in-database embeddings generation

Slick!

If you do not choose to opt-out within 30 days of receiving this notice, Fi will assume your approval to share your CPNI with other Google services and our affiliates.

I assume this is legally questionable?

Should the layperson be concerned with this news?

What reasons?

We'll see a revival of functional programming once we can hammer down the mechanics of AI generated software

I think he mentions this in the podcast. One of the challenging problems they're attempting to tackle is enabling the build to work with standard 120 volt wall sockets. He mentions throttling the system (GPUs specifically) to use less power at cost to performance.

That and setting up the software to work w/ an array of GPUs rather than one seems difficult.

Will Brave ever separate from Google?

I haven't tried self-hosting due to the hesitation around the general drab I've experienced in the past trying trying to host other ML models.

Find a repo. follow the install instructions. What is this weird error? A library issue..? Maybe it's my OS..?

It always seems to be tedious compared to open projects in other domains. Maybe that can't be solved.

Truly.. one of the greatest minds in our ML era. Don't get caught up on the format :)

This is awesome. I have a vision to build self-managed software. This will be a great tool.

Venture capital enables the perpetuation of unsustainable models in favor of eventual scale. This isn’t necessarily a bad thing. The runway to experiment has been critical for any number of businesses. But any good thing taken to excess can become problematic. The same is true with capital

Is it really a good thing to expose retail investors to this much risk?

WEI would be super valuable if it was targeted at corporate network infrastructure.

In those situations, enterprises have the jurisdiction and need to know who is connecting to their network.

Putting a technology like this into a browser seems to only benefit sites that monetize their content...

I 100% agree or at least am just as misinformed as you. Seriously, what is the purpose of this library?

I was hoping to get some genuine answers to this question from this repo but just like the langchain documentation is filled with simple hypothetical scenarios that don't make the library's purpose clear.

Okay, I can use an abstraction to come up with a company name. That is not an interesting use case nor does it explain the need for the library.

What are these hypothetical complex systems that necessitate these abstractions and organization framework? I'm genuinely curious.

There is a huge opportunity to integrate large language models w/ these types of platforms!

I agree that the hype has dissipated. I'd probably contribute a large part of that to media cycles though.

Personally, I think the use cases that go beyond chat are going to be the most valuable. Specifically, the ability to produce structured information.

The depth of the subject is much too great for an article. But, there's certainly some interesting discussion to be had around the relationship between people's general work outcomes and a strong gating mechanism evaluated earlier in life.

I really hope we soon see innovation around education. When the entire corpus of human knowledge became available freely - surely we would've seen changes in higher education by now (if they were/are ever going to happen).

This is a perfect problem if you're able to tackle it. I'm super curious on the approach.

YOLO style object detection on a manually created dataset would be my first pass at this problem.

To me, the recent breakthroughs in generative AI seem as big as when the Internet or the smartphone became generally available.

When you pay with credit, their is typically a slight delay in when your loss is realized (e g. whenever your auto pay triggers).

It seems like this delay has huge effects some tangible some not. Dave Ramsey's preaching on only paying with cash certainly covers some of the psychological effects like:

- this is real money leaving my possession

- I can see how much I have left to spend immediately

As someone who switched from Windows to Linux in college, I have no regrets.

All of the rough edges I've encountered (usually) have resulted in a better understanding of how software actually works.