Throwaway here but I used to work in Alexa org at Amazon and was amazed by how big the org was (thousands and thousands of people) considering it doesn't seem to be a big revenue generator I remember constantly hearing about projects other teams were working on thinking "why would anyone use that" or "how would that ever make money" Just trying to shoehorn alexa into as many domains as possible It was like empire bui…
Alexa is in millions of households and Amazon is losing billions
601–610 of 871 posts
Re: Alexa is in millions of households and Amazon is losing billions
#602Earlier quoted context omitted.
Not sure why you're downvoted. There is material difference in prices of electronics due to whatever reason (taxes, warranties, low scale, etc) to the degree that my European friends often purchase laptops and iPhones when they travel to US.
Do those products have worldwide warranties? Often not. That is a real issue with buying electronics outside your home country.
Re: Alexa is in millions of households and Amazon is losing billions
#603Re: Alexa is in millions of households and Amazon is losing billions
#604Earlier quoted context omitted.
> It’s a slog to do the “which one of these is least a ripoff” calculation, at the grocery store Where I live price labels must carry the per liter or the per kilo price too. It's printed with a smaller font size but it's visible and it's often the only price I look at.
The unit price labels here in the US are hilariously messed up. The items you want to compare will be say, a pack of 6 granola bars or a 10 pack. The unit prices will be like: #1: $0.37 / ounce #2: $6.50 / pound or even: #3: $1.82 / each (a price per item) This will even be common within the same exact brand too. The stores seem so bad at picking a consistent unit for a product category that it seems malicious, but i…
Non-metric units are just icing on the cake.
Re: Alexa is in millions of households and Amazon is losing billions
#605Earlier quoted context omitted.
Interesting. How much would one have to pay you to stop using it? Wondering what the use cases are (replace Google? Stack Overflow? etc).
I'd have to be paid probably at least $200/mo to stop using LLMs, they are a massive productivity boost for new projects and can replace google for simple things. As an example of replacing google (since I was writing C++ today and I haven't used it in ages): me: is there an issue with passing "hello" to std::string claude: Passing "hello" to a function expecting std::string is actually not an issue. C++ allows impli…
After you have taken any kind of C++ course, like learncpp.com, you dont have to look this stuff up anymore either.
So if LLMs are worth 200$/month to you, I would be very interested to see what actually learning the tools you use would be worth.
Re: Alexa is in millions of households and Amazon is losing billions
#606Earlier quoted context omitted.
I can't come up with a single item that I would blind order. At most, I could ask Alexa to add an item to a shopping list and then verify the list and place the order myself.
I would love to order a bunch of things - toilet paper, kitchen paper, toothpaste, mayonnaise, sriracha, foil, dishwasher and laundry tabs. Anything that isn’t a weekly purchase that I find myself out of when I want to use it. The problem isn’t that I don’t want to blind order it, it’s that I don’t trust blind ordering it from Amazon.
Re: Alexa is in millions of households and Amazon is losing billions
#607Do people use alexa for anything else but controlling audio? That is the only usage I have witnessed in housholds that had it.
Re: Alexa is in millions of households and Amazon is losing billions
#608Earlier quoted context omitted.
Of the precious little Alexa can do reliably announcing that dinner is ready without having to yell is quite useful.
How big is your house? Just looking to understand under what circumstances what you said might be a good idea. Where I live it would be silly.
Re: Alexa is in millions of households and Amazon is losing billions
#609Earlier quoted context omitted.
> It’s a slog to do the “which one of these is least a ripoff” calculation, at the grocery store Where I live price labels must carry the per liter or the per kilo price too. It's printed with a smaller font size but it's visible and it's often the only price I look at.
The unit price labels here in the US are hilariously messed up. The items you want to compare will be say, a pack of 6 granola bars or a 10 pack. The unit prices will be like: #1: $0.37 / ounce #2: $6.50 / pound or even: #3: $1.82 / each (a price per item) This will even be common within the same exact brand too. The stores seem so bad at picking a consistent unit for a product category that it seems malicious, but i…
Re: Alexa is in millions of households and Amazon is losing billions
#610Earlier quoted context omitted.
Throwaway, used to work at the NLU unit of Alexa about 5 years ago. There is some ML going on but as with all ML projects I have worked on people want control. This means you add rules for the "important" stuff. You also add test cases to make sure the ML works. But if you already have those test cases, why not just match on them directly? There are also advanced techniques for generating examples (FST for example).…
Is this not actually kind of powerful? Having linguists write up a bunch of rules seems a lot more predictable than "rolling a bunch of dice and hoping that some LLM spits out a coherent set of steps". It feels very fractal but on the other hand if Alexa has only a specific gamut of responses it's not exactly a limitless state space right? Very curious about how those rules look like though
The problem is the space of possible commands is waaaaay bigger than the space of commands you can manually handle, which means if you just randomly try stuff 95% of the time it won't work. Users learn that very quickly and end up sticking to the few commands they know work.
The one exception is "search" queries - "how tall is Everest" and so on, but that only really works well on Google's platform because they've done all the work for that already.
Contrast that with LLMs which basically at least understand everything you're asking of them. If you give them a simple API to carry out actions they can do really complex commands like "send a WhatsApp to my wife telling her how when I'll get home if I start cycling in 10 minutes". That's impossible without LLMs but pretty trivial with them.
Obviously the downside is they are prone to bullshitting and might do completely the wrong thing.