subygan

completely anecdotal. my earliest childhood memories in india were of asthma, sinusitis and constant cold. this was consistent until i was 26 and since then I've moved to the US and lived in New York with 0 days of sinusitis or breathing trouble. and i fallback to trouble breathing when i go home again.

even New York is so much more better than Bangalore or my small village in the middle of coffee farms

you're right. it coincidentally solves cancer and infectious diseases as well.

the ultimate utopia

a lot of life is transactional and businesslike (even if at times it doesn't feel that way.)

and just like in business, a lot of opportunities and outcomes are depended on being in the right place right time & knowing the right person.

hard work tips the balance a little bit. not much

lovable is not aimed at devs though

wouldve been nice to see a video of what theyre talking about. words are so ambiguous in what the product is

Ai is a tool in the same way an employee is a tool. very unlike a hammer.

is it not open assembly?

open weight models are better than open-assembly binaries (as you put it) because models are grown like plants, you can shape the open-weight model in a direction you want by feeding it more data and compute (aka finetuning).

which is something that is impossible in a binary.

embrace the new paradigm and it's tradeoffs. without scoffing at semantics and criticising from an armchair.

we're calling it openweights now. what is a logical push to have a completely new paradigm be compared with traditional software binaries?

do you have a $100 mil worth of compute to train your own version of this?

I don't understand what gripe people have with open weight models and wanting it to be purely open source. the training dataset is only going to be a copyrighted set of contents you don't want to touch with a 10 feet pole. let alone have a publicly traded company host it for you, even if they internally are training on it.

these things will be evaluated based on the net benefit such a service would provide. more people having affordable commute to dallas would end up increasing the net benefit and value that dallas has.

cities grow in capabilities in direct proportion to the number of people who have access to it and their capabilities. it might be cheaper labour leading to better roads, plumbing etc

in the absence of a stalling signal. it's better to assume exponential and work backwards than hope the next bottleneck is impossible.

this is one of the common conundrums of life. but for someone living in the grey, this stuff might push them over the edge to be great. that is the point of this article

And the hidden, trial and error runs tweaking the hyperparams. guaranteed, it cost >$5000 in training runs alone.

This does not really work well, if you don't know the complexity of the problem ahead of time and ensure all future conversations go to the same model.

Else, you break the cache by doing a round robin of the same conversation across different models. Likely you'll end up paying more than what it would've cost with a cache aware system

don't worry, we can get another swarm of random monkeys typing and decide what ones to eliminate .

tijebs gi yo,

hugged to death at 8 points and 20 minutes in frontpage.

Please use AI (or anything) to fix your site.

For thinking machines, they provide super simple finetuning APIs.

if it is their model, they can have more lower level integrations for that. Thinking machines might be the only large lab in the US to have business interest aligned with open sourcing strong models that are customizable.

I wasn't around for the Search engine boom. Did people write about how search engines are making people dumber?

This is true of truly skillful people in their craft as well.

I know people, who initialize all required variables and write the logic after. which used to feel bonkers to me until I realized, they've done enough practice and memorization to be able to figure what they would need 10 steps down the line.

this does show that, models have a better model of the task and the expected end state.

I like the narrative but the key point

> A battle of two narratives > Build wealth before AI obviates our skills > Build skills, agency, taste, judgement

both narratives are portrayed as being odds with each other but, I can't come up with a single "build wealth" scenario that doesn't involve building skills, agency, taste and judgement.

what am I missing ?

What about human written software makes it more reliable than LLM written software?

is it the craftsmanship, or the deliberate decision making of industry veterans?

almost all platforms are like this now. Every active player in each of these platform is trying to get the most eyeballs in their content / profile. and the silent scrollers are not contributing anything else anyway.

We've societally come to the consensus that, we want to reward a race to the bottom slop. passive scrollers by not doing anything about it, active posters by contributing to it.

but there is no way else to win in this game.

A friend of mine writes the most human curated thoughtful newsletter about AI, spending 100 hours. and maybe 200 people know of its existence.

this is awesome! incredibly useful

Is this a submarine article[1] by instacart to sell their consumers data? feels like a glowing review of the data. And why would papa jhons accept to be a part of this campaign?

[1] https://www.paulgraham.com/submarine.html

Why can't this be sent as a notification in my phone?

As horrible as it sounds.

For the median user, It really is impossible to have an alternative to instagram / whatsapp / facebook. It is so easy to live in a bubble and say I'll host my own things. but a totally different thing to have a functioning network effects machine.

i don't at all think it's that obvious / easy.

i was taught cursive in 2nd grade. and my handwriting is gobsmackingly horrible. coming back to stuff I've written after I've forgotten the context, makes it impossible for me to understand what I've written.

and it's not for lack of trying. I spent almost every summer till 10th grade, practicing writing 30 pages a day. and still it gets reset to my horrible hand writing in weeks after school start. at this point, i just consider myself hand writing challenged.

i cannot tell you how much happy i am that, computers have made handwritten exams obsolete.

this is just false.

by any meaningful measure of intelligence. the latest models are much smarter than the bulk of the population.

how would you define intelligence?

I'm a first time father of an 80 day old.

And I can gladly say, these have been the most incredible days of my life. I want to be more ambitious and more active for my kid, in ways I hadn't thought before.

Happy Father's day to all.