nprateem

15 ms·

241.
by nprateem 1y ago
Yeah but it's XML not pydantic which means it doesn't play well with failovers to other providers. It would be tolerable if Anthropic didn't have such abysmal API uptime but at this point no way will I use them for my SaaS.
242.
by nprateem 1y ago
Meh, Anthropic are dead to me until they have structured output.
243.
by nprateem 1y ago
Well I wasn't suggesting mandatory attendance to every event, but I think if someone doesn't show up for say 2 months their space should be freed up for someone else.
244.
by nprateem 1y ago
Lol. I'm sure they'll be gutted you won't be joining. I know plenty of single guys who'd like this sort of club. This definitely emphasises the importance of the filter event...
245.
by nprateem 1y ago
No, you need regular events, same place twice a month. That's the killer for social clubs like this. Just have a regular date in the diary and if people are free they can drop in. Then do the more unique events on a different day (and
246.
by nprateem 1y ago
Weird survey questions. How about "none of the above", "yeah maybe it depends", etc. I don't know what you hope to achieve with such a limited set of answers.
247.
by nprateem 1y ago
I think we're hoping for more than the 'infinite monkeys bashing out semantically correct code' approach.
248.
by nprateem 1y ago
The editor has his finger on the pulsar
249.
by nprateem 1y ago
No one cares about your 'secrets' as much as you think. They're only potentially valuable if you're doing unpatented research or they can tie them back to you as an individual. The rest is paranoia. Having said that, I&#
250.
by nprateem 1y ago
Is it in a Faraday cage too or do you just confiscate their phones. Or do you naively believe they aren't just using AI on their phones?
251.
by nprateem 1y ago
AGI will be v6, so this will get them there faster.
252.
by nprateem 1y ago
I posted it earlier. Anthropic: You're killing yourselves by not supporting structured responses. I literally don't care how good the model is if I have to maintain 2 versions of the prompts, one for you and one for my fallbacks (
253.
by nprateem 1y ago
Their new 'thoughts' smack of "reticulating splines"
254.
by nprateem 1y ago
Finally an AI story here so I can post about leaving Anthropic. Sorry guys, yes, Claude is the best model, but your lack of support for structured responses left me no choice. I had been using Claude in my Saas, but the API was so unreliabl
255.
by nprateem 1y ago
Stability is conspicuously absent from the imagen benchmarks. I assume that means it's significantly better
256.
by nprateem 1y ago
There's s knack to seeing auras. You need to soften your focus and kind of look more with peripheral vision than in the centre. It's pretty easy to see the layer closest to the body. It's kind of like a bright outline about 1
257.
by nprateem 1y ago
LOL did you even read the page you've linked to? OK great, I can get a refund if I move into a care home. I think you're missing the point.
258.
by nprateem 1y ago
It's a way to pay the annual fee monthly. You can't cancel whenever you want.
259.
by nprateem 1y ago
Like I said. Free market, who cares. But not for a publicly funded organisation. And if one group of talent moves on it'll make space for the next generation.
260.
by nprateem 1y ago
All media is propaganda. At least it's not the made up right wing shit that runs America. 1 downvote explained.
261.
by nprateem 1y ago
Gary Lineker, the radio presenters, news readers. Fine if you're commercial but in no world is it justified to pay the likes of Huw Edwards £600k to read the news, then whinge about not having enough money as a public broadcaster. &quo
262.
by nprateem 1y ago
I like the beeb, but I don't watch it since I can only buy an annual licence and if they can afford to pay millions on individual presenters they don't need my money. Bring in monthly subs and cap presenter salaries and I'll
263.
by nprateem 1y ago
It can remove them itself after it's written the code, but basically seems incapable of writing clean code to begin with, unlike Claude
264.
by nprateem 1y ago
Duh why didn't I think of that. Oh wait it was the first thing I tried and it makes NO difference
265.
by nprateem 1y ago
Maybe this one can stop writing a fucking essay in code comments. I'm now no longer surprised just how consistently all the gemini models overcomplicate coding challenges or just plain get them wrong. Claude is just consistently spot o
266.
by nprateem 1y ago
It was just obviously worse than using the anthropic website. That was the only explanation for why it was so bad. They could offer it free because it was stupid even if the same version (maybe less resources). Or maybe I was just unlucky b
267.
by nprateem 1y ago
It makes perfect sense if the market is cheaper.
268.
by nprateem 1y ago
When I tried Claude in copilot it was so obviously crippled as to be useless. I deleted copilot and never went back.
269.
by nprateem 1y ago
It's not an illusion if he can threaten to disable access, etc. which of course he could, just like China.
270.
by nprateem 1y ago
They mean for themselves.

More ›