xienze

12 ms·

271.
by xienze 6mo ago
> Because a machine can never take accountability. A business leader can though. > Compared to another software engineer who directed prompts to expand the system and generate extra revenue streams. I think you're missing the poi
272.
by xienze 6mo ago
> How long do you think it'll take for the AI trend to mostly automate the parts of your job that still make you excited? Yeah, no one ever thinks beyond "whoa, how cool, I cloned Slack in 15 minutes!" Personally, the thin
273.
by xienze 6mo ago
> domain experts will be fine But I don't see how this holds up to even the slightest amount of scrutiny. We're literally training LLMs to BE domain experts.
274.
by xienze 6mo ago
> Hence why I've been quite bullish on software engineering (but not coding). You can easy set up 1) and 2) on contrived or sandboxed coding tasks but then 3) expands and dominates the rest of the role. Why can't LLMs and agent
275.
by xienze 6mo ago
> We don't know what models they use, how system prompt changes or what are the actual rate limits (Yet Anthropic will become 1 trillion dollars company in a moment). Not just that, but there’s really no way to come to an objective
276.
by xienze 6mo ago
> I've always felt if you really want to impact election fraud, tax the hell out of votes. Like $1,000/vote. You don’t even have to go that far. $10 and a trip to the DMV is apparently an insurmountable barrier.
277.
by xienze 6mo ago
> Wondering which of the 8 providers has your show is lame. Yep, but there is a solution. Buy physical media and rip it. You get basically the best quality available _and_ a backup at the same time. You don't have to resort to pirac
278.
by xienze 6mo ago
> Anecdotally, my kids' schools (sample size two, both high school) are quite anti-AI in the classroom. Well, it's still early days. Wait until we truly are in a "learn AI skills or be left in the dust" world and AI w
279.
by xienze 6mo ago
It's not called a uniparty for nothing. Vote red, vote blue, we're all gonna end up in the same place eventually, the only difference is the timeline (pretty interesting that the first states pushing this stuff are California, Col
280.
by xienze 7mo ago
Yeah part of me thinks the reason we know all their claims are bullshit is because you’d have to be pretty dense to think that you could promise eliminate >50% of jobs in many high value sectors within 12-18 months and _not_ expect to cr
281.
by xienze 7mo ago
But that's my point. LLMs generate good prose "most of the time", certainly better than most people are capable of doing. Yet we frequently react with disgust when we see tell-tale signs of LLM-generated text in articles. Why
282.
by xienze 7mo ago
Oh, you'll still need a human to give the initial prompt, like say "Write me a Notion clone." But after that, what value is the human developer really providing? Expertise? Why can't LLMs advance sufficiently to cover th
283.
by xienze 7mo ago
Yeah yeah. But LLMs certainly have been embraced by a large number of developers. Many of whom I've observed react with disgust when they see "not X, but Y" or emdashes in an article. But when it comes to code, "wow this
284.
by xienze 7mo ago
I find it amusing that software developers have no issue with having an LLM churn out slop code but have such a visceral reaction to slop articles.
285.
by xienze 7mo ago
> The field has basically aligned fully in a way it hadn't before that this is an urgent concern. AKA “we want more funding.”
286.
by xienze 7mo ago
> Write automated tests to make sure that inputs and outputs are ok Write them by hand or generate them and check them in? You can’t escape the non-determinism inherent in LLMs. Eventually something has to be locked in place, be it the a
287.
by xienze 7mo ago
> I think that the prompt is the thing that should be PR'd at this point, because ultimately the spec is what's important. The fundamental problem there is the code generation step is non-deterministic. You might make a two sen
288.
by xienze 7mo ago
> At some point I think it makes more sense to fine tune the prompts to get increasingly more specific and just regenerate the the code based on that spec, and store that in Git. Generating code using a non-deterministic code generator i
289.
by xienze 7mo ago
> I suspect if people saw the handwritten code of many, many, many products that they used every day they would be shocked. Absolutely. The difference is that the amount of bad code that could be generated had an upper limit on it — how
290.
by xienze 7mo ago
> If it's harder to work with, it's harder to work with, it's not the end of the world. Yeah it just takes longer and makes you miserable in the process. No biggie!
291.
by xienze 7mo ago
> they just move slower so it’s not as attractive of a target. Well, there’s other things. Maven doesn’t allow you to declare “version >= x.y.z” and doesn’t run arbitrary scripts upon pulling dependencies, for one thing. The Java clas
292.
by xienze 7mo ago
Well, two things. First, “hi” isn’t a good prompt for these thinking models. They’ll have an identity crisis trying to answer it. Stupid, but it’s how it is. Stick to real questions. Second, for the best performance on a Mac you want to use
293.
by xienze 7mo ago
> Microsoft is seriously the worst offender in shoving AI down everyone's throats. The worst, or just ahead of the curve? Because you’re kidding yourself if you don’t think every other AI company or company integrating AI into their
294.
by xienze 7mo ago
> Not even that! This study doesn't even say contamination is causing overestimation. It says that it's possible. From the article: > They found that on average, the gloves imparted about 2,000 false positives per millimeter
295.
by xienze 7mo ago
Yes? Most people don’t live their entire lives in a lab wearing nitrile gloves, so there’s an argument to be made that the concentration of microplastics found in that setting is not reflective of everyday life. So, not that microplastics d
296.
by xienze 7mo ago
That also has pretty poor memory bandwidth. 283GB/s I think.
297.
by xienze 7mo ago
> This isn't a solvable problem without world models. I wish we could use something like a simple DSL rather than English prose to work with these models, in order to have some real precision to describe what we want.
298.
by xienze 7mo ago
To be fair, it’s not really their fault that there are people who want to treat work that normally would be considered a way to pick up a few bucks during free time as a full time career. Sure, go ahead and make fast food delivery a highly
299.
by xienze 7mo ago
> This isn't any different than the "person who wrote it already doesn't work here any more". Yeah but that takes years to play out. Now developers are cranking out thousands of lines of “he doesn’t work here anymore”
300.
by xienze 7mo ago
Yeah I had a similar experience one time. Which is why I laugh when people suggest Anthropic is profitable. Sure, maybe if everyone does API pricing. Which they won’t because it’s so damn expensive. Another way to think about it is API pric

More ›