mpeg

Now please do one to replace the MBA executive that gets hired to restructure the engineering team, asking for a friend.

Spanish is certainly an easy language to learn, but it’s a very hard language to master.

I’m a native Spanish speaker and I find English poetry much easier to understand, English is more rigid and while technically it might have a larger vocabulary you can do less with it, while the flexibility of Spanish in terms of word order and suffixes makes it possible to craft some really labyrinthine verses

to be fair, the title is very misleading, it took me a minute to understand what they meant

The title is a bit confusing, they're not using ciphertext for inference – they're passing ciphertext around in cases where an agent calls into another agent without exposing the plaintext to the end-user

Inference is still done in plaintext after this multi-agent message gets decrypted in the server side

As a counterpoint to all the negative comments here, it’s entirely possible the CTO knows it’s gonna take x time to get your company laptop, which precisely because it’s a regulated industry might mean days/weeks of you waiting around with nothing to do

In that situation it makes sense for you to bring a personal laptop which you can use to access non confidential stuff, you might not write any code in it and might just be given docs to read through

There are reliable ways of differentiating human from cheap, bulk scraping bots.

But if the bot is advanced / expensive enough, it gets a lot harder. Where this product's market sits is in giving a paid way to access content compared to having to spin up bots that run js, from real IP addresses, etc. all of which are more expensive

Yeah, to be fair is not the norm and was mostly due to the AI crypto craze which drove their token price up so they ended up adding very big rewards

Even in LLM jailbreak CTFs I've seen, it ends up feeling like underpaid work when it's sponsored by Microsoft and the prize pool is, say $10k (including stuff like azure credits) considering the salaries AI safety engineers command at big tech!

I saw this thing when it was launched, but IIRC the reward was tiny (like $100?) so it wasn't worth exposing a good prompt for

For comparison, I won a similar prompt injection challenge ran by a crypto company a while back where the total prize pool was over $100k... I didn't win every challenge though, but my team took home around half of that

The problem with good prompt injections is they have a very short half life once they are out in the wild (especially if they work against frontier models)

Data engineers hate this one simple trick

I've used that with companies I consult for, everyone thinks they should do what Google does, so sometimes I'll drop them the "your whole company data fits in my phone/laptop" line to make them understand the (lack of) scale

This is nice, but I don't see much of a difference between just asking claude to write a wrangler config to push to cf pages, which can be extremely simple

  name: my-cool-pages-project   
  pages_build_output_dir: ./html-files/

Yeah I guessed that one right because xylophone player sounded like a trap.

I don't understand how they rank words though, some extremely common words like xenophobia were ranked as high as much more obscure ones.

Yeah I got 75k~ and did something similar ... most of the expert and grandmaster ones had at least 1 or 2 obvious incorrect answers, then it was a 50/50 so I usually went for the thing that sounded either closer to the root of the word or completely left-field

Anything up to expert was obvious

It'd also be a lot less awkward to go through 100 words if it had keyboard shortcuts (1-4 for the words, enter to submit) and if they fixed the layout shift jank

I disagree, a lot of the time people buy "serious BI tools" precisely because they think they need all that power and complexity.

In reality, what most people need is much simpler, a mini app with some curated datasets and simple filters, maybe some AI querying if we want to get fancy. There's some companies out there that work with big data, but for the rest of us small data is ok.

I went with eero and really wish I'd gone with unifi

Apart from the shitty software and basic features either missing or locked behind a monthly cost, the network itself is not bad at all, I get 600-700mbps on wifi throughout the house and have my servers wired on 2.5gbe

But the one thing I really thought I was buying into by choosing an amazon brand was ease when it came to buying upgrades, and yet I ended up having to buy extra hardware (like the wired gateway) from ebay and sellers in the US as amazon does not sell their own hardware everywhere

That seems reasonable, I don't buy NAS for datacenters (just run a modest 80tb one for my home lab) but equivalent rackmount 16-bay ones from other vendors would be more expensive (maybe $5k-6k?) and with less polish.

I think there are unknowns in the calculation:

- What are the margins of Anthropic over their API pricing? Without this, all we're saying is the API is more expensive for heavy usage

- How have their price margins changed over time? I imagine built into their commercial model is the expectation for inference to get cheaper over time

- They have tighter usage windows than they used to, now having both 5-hour and weekly limits, they also seem to experiment with their usage quite often, this probably affects user's average utilisation, do they have any other levers they can pull here? eg how does changing to an 8-hour window affect it, or limiting certain models to API-only usage based on capacity like Fable

I know there is a certain level of subsidised usage built into their subscriptions, the VC-funded company playbook, but I don't think anyone from the outside knows for sure how much it is and I imagine it's lower than most people think, and reducing over time.

This is true for a lot of subscriptions though: some people use 100% of their mobile plan data, some people use 10% – the price accounts for this.

I get that, but the drawbacks outweigh the benefits imho and the b64 overhead is not solved, just deferred.

Say I want to distribute an hmml file as a single file, I'd have to create an html with the embedded js runtime, and then embed the hmml file... as b64, therefore negating any benefits.

I don't understand the advantage of this vs an html file with embedded images, except the latter would work with no runtime.

I like orbstack in theory, but I find it hard to justify a $96/yr license fee for something that has so many open source, free alternatives. As it is, I’d rather use podman or colima

Back in the day I wrote a PoC exploit for my employers app that abused an image upload api by embedding a jar file inside an svg as XXE which then got me RCE. Fun times.

It is, it asks you if you want to continue as opus 4.8… but I was trying precisely to evaluate fable

It's not even very usable... I tried 2 different chats and both eventually got stopped due to the safeguards

One was a piece of code I gave it to improve, it did so and then started writing tests, some of which tested security so the safeguards triggered

Another was one of the cryptography puzzles I use as new model tests, which are hard to oneshot and there's no public solution anywhere, it completely refused to even try to solve it

I figured, I just had a visceral reaction to that specific template as I get so many of them

Random but... please do not use this template on real emails

"Hi [FirstName],

I came across [Company] and was impressed by the work you're doing — especially [specific detail]."

It's a terrible cold email, I get so many of these every week... There's better ways to get someone's attention

I find it quite dishonest as github stars used to be (and maybe still are?) the measure of an open source project popularity, and these big, flashy LLM generated repos seem to always get a bit of attention

Have seen it from jobseekers trying to boost their profile with fake projects, founders trying to make their product more attractive to VCs, consultants trying to advertise their services...

I don't always have time for OSS, but every PR I've ever sent has always been hand written, and tested, and has taken into consideration the project coding style and architecture choices – I don't like this new world where developers can't even be bothered to write the docs.

How does this compare to the classic css-native parallax effect? Before the scroll timeline APIs you'd use the "perspective" css property to create a container where the z plane is n pixels away from the screen, and then position each layer within it at a different z distance using transform: translateZ

That method is GPU accelerated too, so it is performant compared to some js solutions, and has worked well in every browser for around a decade

I like the idea of the scroll-timeline though, just keen to understand what the advantage is for this

This is cool, I have SVT and usually am able to stop an episode if I do slow breathing like that; although sometimes if that doesn’t work the modified reverse valsalva manoeuvre does it every time.