I am a Developer Relations Engineer at Google. Currently I am on the Google Workspace DevRel team and was on the Google Maps Platform before that. Previously I worked at Descartes Labs and the US Geological Survey.
My Debian (KDE) uses just under 1GB on startup. If one is not using animations and things syncing in the background and daemons monitoring file system changes and whatnot, can the stock MacOS memory usage be reduced?
What, in fact, is it doing? I'm of the opinion that RAM not used is RAM wasted, but I prefer that philosophy for application memory, not background OS processes.
> You know what would be good for security: Having physical disconnect switches
Wouldn't those become failure points? Anything mechanical will not only wear, but will be affected by dust, dirt, sand, dead skin cells, body oils, etc.
Light switches do not go with hundreds of thousands of people to the beach, the desert, left in hot cars, rained on, sat on, dropped, pressed against sweaty facts, etc.
I don't think the market of people buying used phones for the purpose of graphene is going to make a dent in profits for Google. It raises resale value maybe by say, $0, considering the price is set by the average consumer
> If anything there fast reduction in value makes them less attractive.
Right. And if you buy a secondhand one you are increasing their value on the secondhand market. Reducing the depreciation increases the value of the brand new phone.
that depends what you consider a healthy resale value, I bought my Pixel 6a with no issues for 100EUR :-) (and not because I care about Google's business, I don't have gapps in my phone, I just like good deals/VFM)
> This is the most fundamental argument that they are not, directly, an intelligence. They are not ever storing new information on a meaningful timescale.
All major LLMs today have a nontrivial context window. Whether or not this constitutes "a meaningful timescale" is application dependant - for me it has been more than adequate.
I also disagree that this has any bearing on whether or not "the machine is intelligent" or whether or not "submarines can swim".
The video mentions that this image was taken during the night time. If the neutrinos do not interact with the entire Earth on their way through, then how do they interact with the sensor?
Thanks. I watched half the video, but couldn't finish it. Now that I've finished it I see that the neutrinos interact with water in the tank.
Thirty interactions on average per day, that's crazy - earlier in the video the host mentions that the cross section of a thumbnail has a billion neutrinos flowing through every second!
The model doesn't even need to be exposed in the UI. Let the user specify "use model foobar-4" or "use a coding model" or "use a middle-tier attorney model".
VIM does this well: no UI, magic incantations to use features.
For many workflows involving real time human interaction, such as voice assistant, this is the most important metric. Very few tasks are as sensitive to quality, once a certain response quality threshold has been achieved, as is the software planning and writing tasks that most HN readers are likely familiar.
The way that voice assistants work even in the age of LLMs are:
Voice —> Speech to Text -> LLM to determine intent -> JSON -> API call -> response -> LLM -> text to speech.
TTFT is irrelevant, you have to process everything through the pipeline before you can generate a response. A fast model is more important than a good model
Source: I do this kind of stuff for call centers. Yes I know modern LLMs don’t go through the voice -> text -> LLM -> text -> voice anymore. But that only works when you don’t have to call external sources
An “intent” is something that a person wants to do - set a timer, get directions, etc.
A “slot” is the variable part of an intent. For instance “I want directions to 555 MockingBird Lane”. Would trigger a Directions intent that required where you are coming from and where you are going. Of course in that case it would assume your location.
Back in the pre LLM days and the way that Siri still works, someone had to manually list all of the different “utterances” that should trigger the intent - “Take me to {x}”,”I want to go to {x}” in every supported language and then had to have follow up phrases if someone just said something like “I need directions” to ask them something like “Where are you trying to go”.
Now you can do that with an LLM and some prompting and the LLM will keep going back and forth until all of the slots are filled and then tell it to create a JSON response when it has all of the information your API needs and you call your API.
This us what a prompt would look like to use a book a flight tool.
Using LLMs for voice assistants is relatively new at scale that’s the difference between Alexa and Alexa+ and Gemini powered Google Assistant and what Apple has been trying to do with Siri for two years.
It’s really just using LLMs for tool calling. It is just call centers were mostly built before the age of LLMs and companies are slow to update
Understood. This overlaps with a side project where I’m getting acceptable (but not polished) results, so trying to do some digging about optimizations. Thanks!
One of my niches is Amazon Connect - the AWS version of Amazon’s internal call center. It uses Amazon Lex for voice to text. Amazon Lex is still the same old intent based system I mentioned. If it doesn’t find an intent, it goes to the “FallbackIntent” and you can get the text transcription from there and feed it into a Lambda and from the Lambda call a Bedrock hosted LLM. I have found that Nova Lite is the fastest LLM. It’s much faster than Anthropic or any of the other hosted ones.
reply