In my world, most multilingual speakers are children of immigrant families, and that transcends socio-economic boundaries. Plenty of immigrants were already wealthy before coming to the SF Bay Area, where we import highly educated, specialized workers. On the other hand, we also import physical laborers who are also generally multilingual but not in the same social class.
Some of these immigrants are very well supported with a strong social network, while others struggle with isolation.
There’s a few polls that have shown most people use AI, but they also dislike it. I’m in that boat, where my company pays for my subscription, and I use it to be productive. But I don’t really feel good about it.
There are plenty of providers of open models that offer very affordable rates. Generally, I recommend looking at OpenRouter since they track various metrics for the various providers.
> Like, do you even understand what you're saying?
That comment is unnecessary and has the effect of making people feel bad.
I think the rationale is that wages are stagnant in comparison to investments (stocks) and costs (inflation). So there's decreased incentive to focus on wages as a form of income, and more incentive to focus on investments.
I've definitely felt this personally, as my income shifts towards investments, my will to work for a wage has decreased. That shift has increased because I've accured more investments, but also because investments have grown kind of ridiculously compared to my wage.
I have a wife. I love my wife. My wife loves me. I cannot return my wife’s love for me at the same amount or manner. She loves me more than I can ever love her. She loves me in ways I can never.
It’s very poetically written and sounds very loving. My simple translation loses a lot of beauty.
I think this is a tree density problem. Most cities have a small number of trees, and they’re almost always over cars. These are trees that line streets and parking lots. Without trees, birds just have telephone poles and wires, which are also over the cars.
In San Francisco, we have a lot of trees on most of our streets, and many parks small and big, all full of trees. This means birds spread themselves out everywhere, not just over cars.
I think the true barrier to getting more trees is that individuals tend not to want to pay for and maintain trees. This includes caring for the tree, trimming it when it gets bigger, and cleaning the pollen, leaves, fruits, and branches that fall.
Cocaine (the powder) is extracted from the coca leaf, which indigenous South Americans have chewed for over 8000 years. While the synthetic drug is insanely addictive, the natural form is still commonly used as a mild stimulant, probably safer than caffeine in coffee. So yes?
Businesses really underestimate how much having a human representative helps customers feel connected to a business. I see it in corporate sales (B2B) where accounts are pretty much tied to the account manager. When the manager leaves, the companies refuse to renew because the account was only good because of the manager.
I think of my favorite businesses I regularly visit and they all have a memorable face to them. I feel more than a consumer. They help me understand the product and guide my decision making. They tell me when my order doesn’t make sense. And they refer me to other places they recommend. Or they tell me my problem is real and a mess, but assure me they’ll fix it.
LLMs will always have the tool call overhead, which I find to be quite expensive (seconds) on most models. Directly using vector databases without the LLM interface gets you a lot of the semantic search ability without the multi-second latency, which is pretty nice for querying documents on a website. E.G. finding relevant pages on a documentation website, showing related pages, etc. Can be applied to GitHub Issues to deduplicate issues, or show existing issues that could match what the user is about to report. There are plenty of places where “cheap and fast” is better and an LLM interface just gets in the way. I think this is a lot of the unsqueezed juice in our industry.
> From a critical perspective, this signals organizational dysfunction. If a company requires 13 people to sign off on a hire, it… implies a fear of making mistakes… The company with 13 rounds was fishing for a reason to say "no”
I realized this at my current job. The decisive interview decision and feedback impressed me. Once on the inside, I could see how the “bias for action” and push for decisiveness permeated the whole company. PRs get approved timely. Meetings push for a conclusion. When someone complains about being stuck, neighbors will offer advice or even a helping hand. I’m so much more productive here than anywhere previously, and I owe it to the culture. They WANT people to succeed. But success comes with risk of failure, so the culture needs to accommodate some failure to allow people to safely take risks.
I’m my interview, I misunderstood the question and presented a solution. The interviewer tried to correct me but I didn’t understand what my mistake was. They encouraged me to just go for it. I eventually realized what they meant, I corrected myself and all of it was a stronger yes signal for them. I push forward, see mistakes, pivot fast, and iterate quickly on feedback.
Interviewers are often unsupportive or looking for a reason to say no. It screams that they’re not really “desparate to hire” and in-fact, may be difficult to work with.
The problem with LLMs using full-text-search is they’re very slow compared to a vector search query. I will admit the results are impressive but often it’s because I kick off an agent query and step away for 5 minutes.
On the other hand, generating and regenerating embeddings for all your documents can be time consuming and costly, depending on how often you need to reindex
I wonder how our mental model of nature will evolve over the next decades. For example, in the early 1900's, the United States had more laws protecting animals from overwork than it did for children. That feels unfathomable in today's United States, where animals are treated more as property than people. Perhaps something similar will happen, where we will understand everything as a "legal entity" that has protections.
While I think I understand your point, there’s probably a few ways to look at this.
One is many products start out pleasing most users, but pivots to enterprise customers because of revenue. Thus, the product shifts heavily towards the enterprise use-case of a few customers at the loss of most small-medium users. Getting more users in this enterprise world means making changes to accommodate special needs and that leads to entropy.
Another new need is to hit next quarters revenue targets, so companies find more juice to squeeze somewhere.
Some of these immigrants are very well supported with a strong social network, while others struggle with isolation.