_kidlike

he did not invent anything AFAIK. but it's the first cohesive work around this (that I've found).

I agree on the first part. the last part is simply wrong. adding engineering principles around AI have measurable outcomes (at least for software development). look at Agent Control Theory from Mark Pollack.

sounds very similar to jte.gg for Java

x402.org

lmao

for the same reason we used ElasticSearch for vector search, because it's just one part of a query. Everything else we already had (filters etc), just stay and can be combined with vector search.

this looks awesome! A very naive question, because I never bothered to learn how it works, but can I embed this in markdown files with ```reladraw, as I can do with ```mermaid? Mermaid gets rendered inside markdown files, in both IntelliJ and Github. I assume both added support for it at some point. But I don't know the mechanics of how this works. I wouldn't be able to use this until then, unfortunately.

that exactly how the Greek word sounds :)

ιστορία, if you wanna copy paste.

and yes, the English word comes from the Greek word!

can we choose read-after-write consistency at query time, if needed?

Do you know if Opus 5.1 is coming and will have improvements in writing style too?

:exploding-head:

I didn't say this is not supported. But it's not as good as streaming from Spotify directly to the amp, and the media decoding happens locally.

Adding the Bluetooth layer adds more layers of encoding/decoding, and generally speaking it has narrower bandwidth.

Also, I know it's a bit niche, but Deezer does not have an equivalent of "Spotify Connect" protocol, which means that Network Integrated Amplifiers (like Onkyo A-50), CANNOT implement a player with good UX. Spotify is now king since they introduced lossless quality which is enabled by default in such amplifiers. The benefit is that there is no local streaming (e.g. Bluetooth) and no latency. The data come from Spotify directly to the amplifier and then your speakers.

So, Deezer that had a audiophile target audience, has effectively lost the game.

JetBrains is a good option - works with any model/agent. Also they have very good semantic search for the codebase, which is exposed via MCP to the agents.

it would be surprising if this wasn't already the case.

So, in all of your software, you have introduced randomness in your retries so that the billions of your clients avoid retry synchronization dances?

Thanks!

ok, then it is indeed clickbait. S3 is still "infrastructure".

Also this implies that nobody tried to build a more performant/efficient system for git for ~13 years? That's crazy...

what does it mean that Origin was built for the era that code moves faster than infrastructure can handle? Is it just clickbait, or there's something fundamentally different about Origin?

in addition, it also has not a great name... the word Oxis in Greek means "sharp", not agile. you can look for "οξύς"

not accessible just by typing another pin on stock Android though.

I had this on a Xiaomi, maybe 10 years ago? Very cool feature! I hope they still do it. I think the wipe feature is also very cool, but not used in this way.

o_O

cholesterol is one of the most crucial molecules for survival. It's used to create and maintain cell membranes, and other critical functions. The vast majority is created by the liver, and there are a lot of regulating signals for how much, which is where usually things go off. Low cholesterol production is an extremely serious condition. The "only" problem with high cholesterol is that it builds up inside artery walls, which eventually may or may not dislodge and may or may not cause a heart attack or an embolic stroke. (I'm not a doctor. I just get medication for high cholesterol and have done some research)

this is not a new system... my father had it 20 years ago, sold for dirt cheap by the local stationary shop (not leather of course).

sunshine + moonlight is the streaming stack. Steam's one is pretty bad, and still only works in X11.

I've set up sunshine + moonlight on Fedora KDE Wayland, and I've played remotely from a different country with unnoticeable input lag (using a gamepad, playing Cyberpunk). The whole thing is even behind WireGuard.

I have very fast internet on both sides, both fiber to the home, with only the tablet running moonlight being on WiFi.

aaahhh...

interesting... I have never seen such a benchmark before, where it forces the model to use a json schema for parsing a document. Sounds a bit counter intuitive to me, but I'm not that deep in the field. I usually just "ask" the model for information that is inside an image pdf. I haven't run any sophisticated benchmarks though...

I assume you mean for coding, for which I would agree. But for OCR Mistral is SOTA :)