jcims

Basically except it's using an Apple service for the backend.

Which will likely get cut off at some point.

There's definitely a level of inauthenticity in the hype, but that's a function of the times we're in.

Revolutionary? With an appropriate harness could could do the same thing with the vast majority of modern large language models.

It's just so much faster and so much cheaper that it feels qualitatively different.

I also think it represents a bit of validation for folks looking for ways to bake models into hardware. Sometimes it's perfectly appropriate to sacrifice good for fast and cheap. I haven't asked Jev to do anything that GPT-3.5 would have likely done worse with.

The fly brain thing on the other hand.

I just had Claude and Jev combine forces last night. I've built a few personal browser extensions in the past and thought it would be fun to copy an experiment I saw on twitter where Jev classifies comments/posts etc as slop or not.

Fed Claude an api key from typesafe and a link to documentation, and within about 10 minutes I had a view of HN that was populated with a little ranking as to sloppiness of each comment.

When your mind has been wired a bit to LLM latency, it feels extremely fast, and for such a subjective rating I think it did a good job.

Feels like it sits in a space between traditional ML classification and the frontier models. I can't think of a 'real' production use case for it in my sphere of influence, but certainly some will. And of course there will be five Jev competitors by the end of the year.

Wonder if it had anything to do with this - https://health.aws.amazon.com/health/status

I hope that one of these days my great grandchildren will be able to go to the Smithsonian and see the beat up and dusty Curiosity rover live and in the proverbial flesh.

We're allowed to have our ceremonies.

Isn’t this the whole concept for (braces) terafab? Reduce iteration cycle time.

>I see 3 kinds of potential users.

3 kinds of users that don't scare the shit out of you.

My buddy works with lots of folks that have gone all in on Harvey. There are other kids of users.

People just blow my mind sometimes. I just love seeing these DIY projects that seem like they apply a collection of skills that would take me 10 years to learn each.

Writeup was very refreshing to read, felt like someone actually wrote it.

I've definitely picked up 'lever', I scrub it every time that I notice I write it but it's in there.

You have to get nearly halfway from the surface of the sun (photospere) to its core before the density of the sun’s material is equivalent to that of the earth.

You don't really say how long you've worked with these folks so I'm guessing here, but in my experience you have to build credibility and trust first.

Focus on the opportunities rather than the problems, connect people doing similar things in different areas and jump in when they are solving problems after the fact. The last part, for me, has been key to building it in my career (30 years in infra/tech/security).

That's exactly what had me scratching my head!!! Feels like it' (sorry) unlocks the boundary of the specifics and lets me work with ideas.

I'm actually still quite bullish on generalists. Specialists advance every front but build the supply lines between them.

In favor of the generalist, I think AI is also quite limited in its scope of how it generalizes. I'm mowing through hundreds of mythos-generated security findings right now for work and while it's amazing that it can build an exploit chain 20 steps deep, it's completely lacking in all of the external layers that render it's speculation moot.

>Quote: "The Overhang consists of the unrealized capital gains of past mathematical creativity, the latent value from connecting the dots in the existing corpus. It is a dividend of canonization. Mathematician X states problem A, mathematician Y crafts concept B, then mathematician Z notices that B trivially solves A and “captures” the social reward.

I've made an entire career out of being 'jack of all trades, master of none'. Being able to synthesize connections from relatively trivial knowledge in a bunch of domains is SOP for many humans as well. I think AI just has deeper knowledge and better pattern matching to make up for it's (at least now) lack of strength in cognition and 'ex nihilo' creativity.

(Which probably isn't 'ex nihilo' at all, and has more to do with the plethora of modalities that humans live in vs. large language models. For example, why do we pick the color red for notating important things and why do we say a schedule 'slips'...these are informed by a shared human experience borne of distinct physical sensation deep in our wiring that LLMs can only infer from what we write.)

I tinkered with gnuradio back when rtl-sdr just took off (2012ish?).

I had no background in dsp or signal processing and found it so opaque as to be unusuable, even though I saw more skilled folks do amazing things with it.

*scratches head*

Might give it another go.

We hired a new guy at work that's leading a specific platform initiative.

In person he's very articulate, able to communicate abstract thoughts clearly, states clear goals and how he's learning about the business to develop a plan.

But dear lord everything he writes in email and slack reads like it's copy paste from chatgpt. It's unnerving.

I've gone through several cycles of religion in my life. In various parts of that spectrum I've wondered why humans invented it. Five decades deep it's becoming more and more evident to me that there is some kind of deeply ingrained need for it in wide swaths of humanity...regardless of what it actually looks like in practice.

First time I saw an electric plane with the cowling off was something like this:

https://www.wired.com/story/magnix-electric-plane-motor/

So much simpler.

Love this idea! Any time I visit a place that specializes in a cuisine I'm not familiar with, I tend to struggle with what to order because I don't want to commit to just one entree. Having a curated tour of tastes, so to speak, targeted to folks exploring the cuisine would be absolutely perfect. Wonderful!

>The idea that it’s actually better for diversity feels like a wild reach, almost a canned talking point.

lol

She's got a photo of a new student from Cameroon pointing at a little laminated cartoon character that she hung on the bulletin board outside of her room.

It's a very effective way to make new students from far flung regions of the world feel seen.

Primary ELL. Could certainly be the case in the secondary schools but yeah, nada for teachers, students or admin.

It did make me ask why I wanted to see them while reading, and if there was any value in just reading the text without the images filling in the blanks (and possibly biasing what I was focusing on).

Concluded that I still want to see the images right around the time the are referenced.

This is incredible to hear! My gf has had similar stories about kids coming into her program with zero English, often with parents that have little to no English, and within a short amount of time becoming conversationally fluent and translating for their mom and dad.

It's pretty amazing.

Eh.

Prior to her current role my gf spent about a decade as the head of a very elite private elementary school (she's been in education for 35 years now). Certainly ten figure families represented in the student body with names you would recognize as such. The school had access to all of the resources they could ask for and they did very well on paper. However, the attitude of the students was qualitatively different (and in my gf's opinion much worse) than those of her current students.

It's fun to imagine a collection of rock star educators huddled around every child just constantly feeding their intellectual curiosity, but the reality is most of those kids just aren't hungry. They live lives of comfortable inconsequentiality which, on average, seems to rob something from human curiosity and drive.

Granted I didn't comb over it but to me the big issue was that it was just a big data dump without any material attempt at meaningful analysis.

That and all the figures were at the bottom which was kind of annoying.

I didn't scour the text but one thing I didn't see as an explanatory consideration in the paper is that the larger companies they talk about having larger adoption are also much more likely to have reasonably robust controls (and consequences) that prevent mixing personal/corporate use of AI tooling. That's my world.

On the other hand my girlfriend is a teacher and for the past year ChatGPT has allowed her to augment her lesson plans with slides/games/etc that have had a very positive effect on student engagement. She also uses it to help with analysis of testing data against her lesson plans and state standards to find places to shore things up as the year progresses. She runs the english language learning department in her school and also uses it to help ensure that there is representation in her materials for the huge mix of cultural backgrounds of her students.

It's truly helped her immensely and it's all been on her personal account. The district has no program or budget to guide or support teachers and admin on using these tools in education.

>Answering 2,500 independent questions is an embarrassingly parallel workload

So is calculating the total time required to answer all of the questions.

Maybe the thing that has changed is the meaning of 2 months.

Love my Logitech free-wheeling mouse wheels! Not having a heavy wheel and detent-free option is a non-starter for me.