Hacker Newsnew | past | comments | ask | show | jobs | submit | more lubujackson's commentslogin

Auto had a big improvement a few weeks ago (around when pricing changed)


If a few weeks is months I would agree I think the change to Auto was 2-3+ months ago when they moved to charging named models and higher limits on Auto.


Give me a break with this "where are the protest songs" stuff. I'm an old fart, but even I know stuff like Childish Gambino's "This Is America", a bunch of Kendrick Lamar songs (not to mention his Super Bowl performance), Beyonce's "Ameriican Requiem", etc.

And let's not forget that protest songs aren't usually promoted by those in power...


It's crazy to think that "This is America" was released 7 years ago.


2001 and 2016 have been unfortunately been very long years thus far


Also this kind of stuff is still happening, look at all the blowback to Bad Bunny performing at the next Super Bowl


The best insight I heard from a founder was to realize most VCs goals aren't to have a successful exit, but to have a successful next round of funding.

It makes sense when you realize the time horizon for successful VC plays is long and VC firms need some metric to determine which people are funding or promoting good bets.

Much of VC behavior makes a lot more sense when you understand this. It's not the firm you are selling to (for mid to large VCs) but your contact there specifically.

Goodhart's Law strikes again.


You can also just freeze it. My wife has had some starter just hibernating like a space traveler in our freezer for months.


The solve is just rejecting the commit with a "clean this up" message as soon as you spot some BS. Trust is earned!


Worth noting this is far from a modern problem. Google "yellow journalism".


Luckily I never got rid of my old CDs. They have been sitting in a cabinet for decades and last Xmas I got my son a portable CD player for $35. They have been exploring all kinds of my old music, which is awesome.

I see it in your photos here - Dookie by Green Day is a big hit with my boys!


The key point is "relegating the reasoning". The real way to think about interfacing with LLMs is "abstraction engineering". You still should fully understand the reasoning behind the code. If you say "make a form that captures X, Y, Z and passes it to this API" you relegate how it accomplishes that goal and everything related to it. Then you look at the code and realize it doesn't handle validation (check the reasoning), so you have it add validation and toasts. But you are now working on a narrower level of abstraction because the bigger goal of "make a user form" has been completed.

Where this gets exhausting is when you assume certain things that you know are necessary but don't want to verify - maybe it let's you submit an email form with no email, or validates password as an email field for some reason, etc. But as LLMs improve their assumptions or you manage context correctly, the scale tips towards this being a useful engineering tool, especially when what you are doing is a well-trodden path.


I find this to be too rosy a story about using agentic coding to add to a codebase. In my experience, miss a small detail about the code and the agent may can go out of control creating a whole new series of errors that you wouldn’t have had to fix. And even if you don’t miss a detail, the agent eventually forgets because of the limited context window.

This is why I’ve constrained my use of AI agents to mostly “read-only and explain” use cases, but I have very strict conditions for letting it write. In any case, whatever productivity gains you supposedly “get” for its write scenarios, you should be subtracting your expenses to fix its output later and/or payments made for a larger context window or better reasoning. It’s usually not worth the trouble to me when I have plenty of experience and knowledge to draw from and can write the code as it should be myself.


On some vectors, VR for instance. I wouldn't be surprised if a custom LLM isn't already in the works goes beyond the AI girlfriend thing.


There are a dozen firms offering LLM "chatters" for OnlyFans creators.


Speaking of the "landline" thing for kids that is mentioned at the start of the post, this product is making the rounds amongst my kid's class (3rd grade). Not sure if it will catch on, but seems far too pricey for my taste, though I like the concept: https://tincan.kids/

I think it is good to separate the nostalgia from the actual valuable nugget you want to revive. Nostalgia is great for marketing but parsing the missing nugget is the important part.

I have hundreds of CDs I never got rid of and last Christmas I got my son a cheap CD player. Yes, he could have infinite music through Spotify, but what I wanted to give him was that sense of control over music. The physical element has value, which has been appreciated for a while - a lot of that comes from the purposeful interaction required to select, set up and play the music. To listen to entire albums instead of individual songs. An avenue to explore music you only sort of are interested in but give more time because of switching costs.

But more specifically, I remembered the feeling of being a kid and having my own cassette player, walking around with it and bringing music with me. It was one of the first things I owned that could modify my space and change my mood and affect those around me in a positive way. That is a powerful concept when you are little!

I think the missing element of the "old web" is having that sense of control and influence. Not huge control or huge influence, but self-directed and with some friction. Sometimes, the friction is the most important part!


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: