HN user
Software quality
Be honest though, what’s the difference between this and a good system prompt in Claude to be a language tutor?
I worked for an Edtech startup before that had a novel approach to mastery (as opposed to Anki) and it’s not something you can put in a prompt.
Babbel is better than Duolingo!
uv has a lot more perks! It makes distributing python tooling easier too
DO will sign off in minutes (or did for me)
Imagine if they were cheaper and easier to manufacture
Soon we will have low carbon developers
Yeah that’s what I meant, apologies if unclear
Go is definitely better, but with uv you can install all dependencies including python with only curl
Who?
The author is an early CS bachelors student so… they might still learn this in school
Fwiw, Curry is 30 years old! It looks newer than it is fr the site
Why’d they put up fake roles though?
Tbf, saying they did cut “some billions” as a group of unpaid or low paid 20-somethings is fantastic returns
My wife earned that as a postdoc at Stanford, but as a PhD in Oxford earned £14k per year. $55k as a PhD is really really good.
Norway has an insane sovereign wealth fund worth 3x GDP that they can only withdraw 3% from per year and that oil sales go into (as they’re basically self sufficient on hydroelectricity), they own 1.5% of all global equities. Their pension system is probably the strongest in the world.
How do you guys do coding assessments nowadays with AI?
I don’t mind if applicants use it in our tech round but if they do I question them on the generated code and potential performance or design issues (if I spot any) - but not sure if it’s the best approach (I mostly hire SDETs so do a ‘easy’ dev round with a few easy/very easy leet code questions that don’t require prep)
I used a Knights and Knaves puzzle generator last month to test 4o / Claude 3.5 and all failed on novel puzzles
Screw OpenAI, they scrape us without issues so someone scraped them. No issues with this.
This is from lean protein, add fat and your body is okay. You need either fat, carbs, or both. But you can’t just run on protein.
Great comment on the medical site:
RE: "6 to 9 lb of cheese, sticks of butter, and additional fat incorporated into his daily hamburgers"
This implies a daily intake of hamburgers (presumably ground meat only if it's a strict carnivore diet) with 6-9 pounds of butter and cheese extra per day? Is that time frame correct? Is there an approximate daily kcal for that? A pound of butter is 3200 calories, while a pound of mozzarella is 1200.
What sort of edit is that?
They bought between 10k and 50k of them before the US restrictions came into place. Sounds like DeepSeek gets to use them for training, as they were profitable (could still be, not sure).
Having lived in the US, UK, and Sweden, I prefer Sweden for everything except my friend group!
As a European I really don’t see the difference between US and Chinese tech right now - the last week from Trump has made me feel more threatened from the US than I ever have been by China (Greenland, living in a Nordic country with treaties to defend it).
I appreciate China has censorship, but the US is going that way too (recent “issues” for search terms). Might be different scales now, but I think it’ll happen. I don’t care as much if a Chinese company wins the LLM space than I did last year.
They sold lots of animals, but they found no viruses in the animals themselves right? Sampling found it in the environment, but not in the animals?
I might be wrong but I swear I read they never found cov 2 in an animal that could transmit to humans
How is occams razor pointing you to a fish market where they didn’t find it in any animals that could transmit it, next to a research center working on similar viruses, where people commuted from there via the market, where documents exist pointing out poor security practices, and where the Chinese government restricted access?
Also in one article I read the intel or reading hadn’t changed? Just the reporting is open now
Sounds like phishing? Try go through their website and see
This feels like fluff. You can think a few steps all you like but bugs will creep in, those you can’t think about, those in areas you don’t quite understand, those that require weird sequences of events.
This is just an ad, but in principle I agree integration tests can usually bring more value (depending on the system, but in my general experience across many companies of all sizes).