tinyhouse

Are we talking about online chess cheating? Aren't players eventually need to show up and play in-person?

This is so well written. Well done!

Cursor is great but they're all going to cash out and leave SpaceX as soon as they can.

At least on Claude Code it's completely useless. It tells me to run all the commands myself cause it's blocked ("Classifier's blocking me again" lol). Just tried now and saw the msg: "There's an issue with the selected model (claude-fable-5[1m]). It may not exist or you may not have access to it. Run /model to pick a different model."

I really have no clue how Anthropic released this thing without doing any real testing. I did use it on claude.ai with no issues; talking just for code.

This is very true and applies for everything in life.

I try to always buy tickets on TickPick when I can (no affiliation). No fees and total prices are often much better than Ticketmaster. But my usecase is almost always buying from resellers. I never up-to-date to buy official tickets.

Drew is clearly a very competent engineer who built an amazing product and company. He was the right person to build it from zero all the way to an IPO, but wasn't the right person to keep scaling it. Dropbox's product vision in the last 10 years was lacking to say the least. Their latest innovative product "Dash" is another flop, like Dropbox Password, Paper, and many others.

False is too strong. Biased information is more appropriate.

OpenAI and Anthropic are both private companies with lots of individual investors such as employees, secondary-market buyers, and so on, who stand to become multi-millionaires. So most of what you read about them here is probably colored by someone's financial interests. Not that it's gonna make a difference, but people are just being people.

Looks nice but seems overkill to me to run a Go server to sync with a telegram bot to authenticate. Maybe I don't fully understand the use case.

They own the best models and will probably keep owning the best models for a while. They have much more compute now and more data to keep improving their models on many tasks. Open source won't close the gap in 6 months. They are also trying to block other companies from distilling their models [0].

[0] https://www.anthropic.com/news/detecting-and-preventing-dist...

Pi is my preferred coding agent and I'm happy it lands in good hands. At this point the only reason to use CC is access to Opus. I think open source will end up winning the coding agents race among developers at least. We just need a 30B size model with Opus like capabilities for coding. Frontier labs will dominate the Desktop / Web.

I highly recommend everyone to use Pi - it's simpler and better harness. The only tricky part is that moving forward you cannot use the Claude subscription to access Opus. But for many tasks there are enough alternatives.

I really started to like Pi. That's unfortunate that I won't be able to use it with Opus (way too expensive without a subscription). I'm optimistic that open source coding models will be able to keep up. AI is too important, we're shooting ourselves in the foot if we don't adopt open source tools and models. The more adoption the better it will become.

A lot of these ideas Dspy and RLM (from the same people IIRC) are more marketing than solving a real problem.

That's going to happen in all of big tech (already happening at Amazon and Microsoft). These companies have too many employees. It was never really justified and with AI even more so. I've been in big tech and directors often tell everyone to hire when they can rather when they need. For example, if they know a hiring freeze is coming, they will try to hire as many people as they can before it happens. It's rare to find people in big tech where their incentives align with the company. (and the blame is not always on the people themselves)

As for Meta, I give Mark credit for trying, even if he failed so far with all the VR stuff. The main disappointment is about Llama cause it's clearly an execution problem. With Meta's investments in AI throughout the years, not being able to compete with Anthropic and OpenAI is a big failure.

Israel and the US completely control their airspace and Iran's entire navy got demolished. I think the US prefers not to got too far as they prefer to keep the negotiation talks open. According to reports they asked Israel not to target energy for example.

Less people visit the US because it's do damn expensive. That's the biggest reason for most people. Most people don't have any principles, they go where they can afford. Last year I was in NYC and Miami beach and was shocked how expensive everything was. (I know these are expensive places but that's where most tourists go - they don't visit Kansas)

AI made juniors without potential useless, not all juniors.

Yes, they depend on AWS for compute and Amazon also owns a big chunk of Anthropic (it used to be close to 30%, probably less now with the recent raises). I think it's a good partnership since for the most part they focus on different things and I don't see Anthropic going after AWS - they are an AI company first and foremost. Amazon has their own AI stuff for enterprise but no one uses it so I don't think they take it seriously. They know they cannot compete here.

I think that OpenAI and Microsoft is a more challenging partnership with much more overlap.

Openai is just playing catchup at this point, they completely lost thier way in my view.

Anthropic on the other hand is very capable and given the success of claude code and cowork, I think they will maintain their lead across knowledge work for a long time just by having the best data to keep improving their models and everything around. It's also the hottest tech conpany rn, like Google were back in the day.

If I need to bet on two companies that will win the AI race in the west, it's Anthropic and Google. Google on the consumer side mostly and Anthropic in enterprise. OpenAI will probably IPO soon to shift the risk to the public.

I think the article has some truth but the author also ignores something important. Yes, subscription costs are going down. But there's a big difference between consumer and enterprise. Everyone needs to build fast now. A company cannot get distracted by building capabilities in-house that are not core to their product. This was true yesterday and will be true tomorrow. That means they will keep paying for quality solutions and not settle for sub-par solutions just because someone made them for free (there was always an open source solution available long before AI entered the scene). I may argue that not settling is even more important now that moving fast is key.

For a company, paying $10K a year for a quality service, that's a no-brainer. Most companies spend that money on alcohol in company onsites. However, if you're charging really high prices (the Datadogs of the world), then you're going to face tougher competition from cheaper alternatives that might be as good as you, and when companies need to cut costs, which they often do, you'll be in trouble.

I think what it means to many software companies is that prices will significantly go down on average but the median might not see significant decrease. Companies will be smaller and more lean, hiring less people in general (not just engineers!). There will be more companies out there, so hopefully it will even out.

Last thing is that every product will have too many options to choose from. This has been the reality actually for a long time and going to get much worse. How you market and brand your product and acquire customers will become more difficult than ever.

This is great. AI is too revolutionary to be in control of three closed models / companies. The more the merrier.

(I know this is not a new model but it's not just about the model, it's about the entire ecosystem)

fyi, I just signed up and the confirmation email went to my spam folder (Gmail).

How is this different than Celery and the like?

There's a balance between coding by hand and vibe coding that is important. The less you understand the code, the more boring maintaining the software becomes. It's OK for throw away code, but not for serious open source projects. Use it as a powerful tool rather than your replacement.

I'm not a big fan of this solution since it rewards people who knowingly did something that is illegal. It also allows businesses to take advantage of these people, unless you decide to give them legal status immediately. However, I agree with you that getting the balance right is really hard and that deporting people, esp families with kids who grew up here and did nothing wrong, is very problematic.

Well, if a Korean car factory worker live and work illegally in the country, then it makes total sense to remove them, regardless if they are serial killers or not. A company shouldn't even hire anyone who is not eligible to work legally in the country. There are laws that need to be followed like everything else.

It sounds like you're saying that you want the country to have open borders so that everyone can come live and work here given they pass some basic checks (no criminal history for example). I am not saying that is wrong, but that's not how pretty much every country in the world operates.

I don't have a horse in this race, but I do have a question. If you don't deport illegal immigrants, why not just open the border to everyone to come in? (let's ignore criminal records, etc for this exercise). What's the point of not letting people in but then if they manage to come in illegally, assume it's all good and they can stay?

Well, software is measured over time. The devil is always in the details.