resters

I would think so, I have been thinking recently that I will stop using it based on the decline in quality. It's really frustrating.

Sorry to break it to you, so are nearly all middle class jobs.

I drove Uber Eats for one day a few years ago (as an experiment) and it was way harder than I had imagined. Seemingly no buildings have numbers on them, and it took tremendous effort to find which door to knock on in most apartment complexes, and the tips were lousy and the pay was horrible.

I agree, I think the job is very difficult and does not pay well and so it's hard to find people who want to do it. I'm also aware that I could pay an assistant $100 to shop for me, and the reason I choose Instacart is because the service is not all that expensive.

While it could be a case of capitalist greed, it could also be a case of strategic misalignment of incentives for workers (too little pay, insufficient incentive to keep the job, etc.)

Note that I said I have been a customer for a long time, not that I know how well the company is managed or whether it understands incentives for its workers enough to thrive. This news about acquisition suggests perhaps its model was not working.

It could also be a casualty of the immigration crackdown on labor availability, or a consequence of inflation on tips (when the groceries already seem overpriced bc of inflation, maybe tips are lower), etc.

I'd rather occasionally get a rotten item that has to be refunded than go to the supermarket, collect everything into a cart, and then wait 15 minutes to pay and check out, carry everything home, etc. The service provides a tremendous value, the issue is just the marginal cases where the quality control is poor.

The idea that a word is defined by the company it keeps was decades old. Mikolov’s insight was that you could learn it at scale with a very simple model, and that meaning would fall out as geometry. Obvious in hindsight, foundational in practice.

I've been an instacart customer since it launched in NYC, and I've generally been quite happy, but the quality has been gradually on the decline.

- in the past few years, orders are frequently not received.

- in the past few years, fruit/produce is often covered in mold, rotten, or otherwise unacceptable.

- in the past few years, delivery time estimates are frequently missed.

- in the past few months, customer support got significantly worse and came to resemble the bad attitude and general grudging unhlelpfulness of AirBNB's support team.

I think the main issue has been a big shortage of supply of workers who are willing to do the job and who can do it well. Often times the person delivering the order (and the order itself) smells strongly of cigarette smoke, and increasingly I've had drivers refuse to bring the order to the door, etc.

Add to that the impact of the labor shortage on grocery stores that makes stocking less reliable and results in no availability of bags, greater chances of rotten produce, etc.

Instacart does stand behind it service and I have received full refunds for the non-delivered orders and the rotten produce, but such things must take a great toll on the profitability of the business.

At first I hoped that fears about ICE abuses were overblown, but then I saw video after video of masked, badge-free agents manhandling and throwing grandmas and elderly/weak people to the ground and generally acting like thugs, stalking immigration courts for people trying to comply, doing nazi salutes, etc. It quickly became clear that ICE is largely made up of KKK or proud boy thugs who see it as a chance to brutalize a vulnerable population that they despise.

Now that this data has been leaked, identities can be correlated with facial recognition, gait analysis, and other biometrics to conclusively identify the agents who committed crimes. Much of the video footage shows agents flagrantly breaking the law, abusing power, violating the rights of citizens and noncitizens alike.

Now that his leak has occurred, criminal ICE agents will be brought to justice for many of the crimes committed already, and the data will also inform the prosecution of crimes they will likely commit during the rest of Trump's term in office. Many of the crimes are violations of state law, so there is nothing the executive branch can do to stop the prosecutions.

Maybe Apple is shifting the terminology from display to glass as an homage to fine watch terminology.

<scarcasm>Wait till millions of robotaxis hit the streets, the graph will crater!</sarcasm>

Right, it's the same strategy as when an animal marks territory, seeing if the other animals will resist. They are marking the constitution, many stable norms, the idea that good faith can be assumed, the idea that courts will be obeyed, etc.

Meanwhile the US now has a collapse of due process, concentration camps on domestic soil and abroad, many officials and family members who are suddenly billions of dollars richer, crony firms that are billions richer, and a docile and captive congress that seems unwilling to stop any of it.

It's truly bizarre. I think some of his supporters think that they are all in on the joke and that he doesn't mean half of it. But at this point there is zero evidence good faith, zero evidence of strategy. Just one odd PR stunt after another that someone can day trade on if they know about it a few hours in advance.

It's hard to tell which executive orders are meant as a joke and which are meant as sincere (but horrible) policy perspectives.

I suspect this one was actually meant as a joke, although it is not remotely funny in my opinion.

Someone in Trump's orbit who writes them gets an immature charge out of trolling people.

So far my prediction about Trump have sadly been correct: The US is getting weaker and poorer every day.

Why should we let politicians dictate which workers a company can hire? Not only that we have a politician doing this who has zero technical acumen and zero business success!

see my clarification: https://news.ycombinator.com/item?id=49942421

I think that's why the comment hit home.

There is nothing wrong with practicing craft, but when craft starts to matter more than vision/ambition (as it has for many engineers, even very good ones) there is going to be resistance to any method that de-emphasizes craft and emphasizes vision.

Even things like Ruby on Rails and node.js (in their early days) irritated some software crafspeople for the same reason, the tools felt like the happy path was somehow a mediocre one that was willfully ignorant of craft.

Suddenly, agentic coding has seemed to remove much of the craft from software engineering, and those who have been showing up every day celebrating craft are feeling it like a gut punch.

This is illusory bc craft still matters -- rust is a superb example of how craft matters more than ever in an agentic world, fwiw.

The backlash against agentic coding comes from the split that has already existed among professional software engineers. The ones who are most resistant to it are the ones for whom it has been "just a job", but the ones who love it are the ones who love solving problems. It's now possible for a small team (one person, a few) to solve truly daunting software problems with the help of agentic methods. If you were a 9 to 5er who valued craft but didn't dream about solving problems, AI is a buzz kill.

Not everyone opts for that.

why did they spell it with an A at the beginning?

Frontier models are being used to obtain training data from users. We burn tokens teaching OpenAI how to make a cheaper model that is almost as good. I think the new $500/month pricing strategy is a significant misstep by someone who has clearly not tried Gemini 3.8 Flash or Deepseek 4.1 Flash.

wait till they translate all the critiques dolphins have about human civilization.

Garry has made many gripes about the homeless and other kinds of gripes that do not suggest he has Christian values underlying his views.

It's somewhat amusing to read a defense of the GOP focused on a state over-taxing. Federal over-spending is possible without short-term tax increases b/c the federal government runs a big deficit. If the Federal government had to keep a balanced budget, Trump's policies would have raised everyone's taxes far more than anything the state of California would ever dream of.

(future generations will be taxed for all of Trump's gravy train)

Uh, yes it is quite a lot like it and more invasive/harmful. Trump has also removed many transparency and accountability mechanisms, etc. There is absolutely no reasonable justification for supporting it at this point.

Garry Tan is extremely socially conservative, seemingly unabashedly so. I would not be surprised if Garry had donated to Trump, etc.

Note that whatever it once meant to be "conservative", wanting less government control of markets, etc., today's GOP is the party of economic central planning and limiting economic freedom, and also authoritarian rights violations on US soil.

There is absolutely no excuse for supporting the Trump approach / agenda unless one intentionally wishes to see laws broken and democracy subverted.

Same. Fortunately DeepSeek keeps getting better.

when you think about all of the advancements since Attention / GPT a lot of it has been somewhat more obvious than in other fields, as is typical with the massive flood of innovation that follows a big breakthrough.

Paul likely assumes there will be a sequence of additional papers with the same impact as Attention is all you need, which will spawn a lot of opportunity for a larger group of experts who are conversant enough to advance the field even if they do not themselves create such a major innovation. Not only is this deeply exciting, it is also highly meritocratic as there is still scarcity of the kind of intellect and creativity necessary to swim there.

Machine intelligence might soon surpass it, though, and Deepseek is 100% Chinese mainland educated. Paul's description of building an LLM from scratch is meant as a vague starting point for being an innovator of the highest value aspect of modern AI innovation, not as a specific prescription.

over time greater intelligence will be expressed in smaller and cheaper models. we are still somewhat near the beginning of this bc we are finally starting to understand what makes a model truly intelligent/capable.

With Sol we see openai making the model extremely slow and paranoid about process/ceremony. Sure this is a good guardrail against AI going rogue, but it also sets the stage for companies to charge for 2x, 4x, 8x performance, with 1x being barely tolerable and frankly slower than last year's models (though less error prone).

The irony is that the smarter the model, the more it can be trusted to do with less supervision, so one engineer can manage a team of 20 fable subscriptions more effectively than a team of 3 of last year's model subscriptions.

At this point nobody should assume good faith on Trump's part... not citizens, not allies, not adversaries engaged in negotiation. It turns out "The Art of the Deal" is simply lying.

Yes, thanks for this, Deepseek! (I did start using the Deepseek harness and v4 flash due to the crazy low usage limits on sol and I'm quite pleased to have discovered how capable both of those are -- both have now earned a place in my agentic workflow).