daxfohl

17 ms·

271.
by daxfohl 1y ago
Yeah, but that's not the point I was trying to make. My main point was that traditional thinking is that schools need to fight AI or students won't learn. But I think we'd be better off working with AI instead, customizing ed
272.
by daxfohl 1y ago
Why client side? Doesn't that mean re-uploading the whole context whenever you change something? Seems like storing it server-side and providing an API to swap pieces in and out of context would be a lot more efficient. Or if the LLM i
273.
by daxfohl 1y ago
I think the workflow of translating English to Python, working in Python, and then translating Python to Rust sounds a lot more error prone than just learning Rust if that's your target language. Especially if other people are using di
274.
by daxfohl 1y ago
Seems like that could be a job local LLMs do fairly well soon; not a ton of reasoning, just a basic ability to understand functions and write fairly boilerplate code, but it involves a ton of tokens, especially if you have lots of verbose o
275.
by daxfohl 1y ago
I agree with your main point, though with an asterisk. I don't think concise is necessarily better than long, nor do I think long is better than concise. The thing is, humanity tends to go in cycles. Poems for Babylonians, long epics f
276.
by daxfohl 1y ago
The education system is already a relic of the past. All the points mentioned in the article assumes that a classroom full of students writing five-paragraph essays on The Red Badge of Courage in 12-point TNR with one-inch margins, hastily
277.
by daxfohl 1y ago
Even cars are too much. Give me a Waymo. Or better yet just let me stay home and doom scroll while my life gets delivered to my doorstep. Geesh, doorbell again. Last mile problem? C'mon. Whoever solves the last 30 feet problem is the r
278.
by daxfohl 1y ago
The author may have hit on the answer without realizing it. He's in the gym doing pull-ups. Is his lat strength necessary for some important part of his survival? Highly unlikely. Pre-20th century, when most work was highly physical, w
279.
by daxfohl 1y ago
Its easy to mix up "what seems easier to work with" vs "what's actually easier to work with". Pulling things out to config, splitting into microservices, additional layer of abstraction, rules engines, etc., they se
280.
by daxfohl 1y ago
So whereas previously, repo owners were getting flooded with AI-generated PRs that were complete slop, now they're going to be flooded with PRs that contain actual bugfixes. IDK which problem is worse!
281.
by daxfohl 1y ago
That seems off in my experience, but maybe it's different in different industries. Either way, working around sharding and scale seems like a place where a good AI could help provide recommendations and tradeoffs, so it's disappoi
282.
by daxfohl 1y ago
It'd be nice if it also included the queries for each use case, and had a better understanding of sharding. As is, I gave it a bunch of requirements, it created some tables. Was it sufficient? No idea, it didn't say what the query
283.
by daxfohl 1y ago
Wow, how can you afford offering free AI on the homepage for an OSS project? I can barely afford AI APIs even for my own personal use. Seems like it would drive costs through the roof.
284.
by daxfohl 1y ago
One could also say the opposite. It's not abstract at all, just a set of rules and their implications. Plausibly the least abstract thing there is. On the other hand, two cookies plus three cookies, what even is a cookie? What if they&
285.
by daxfohl 1y ago
Eh, not so sure about any of this. There's also the possibility that math gets so easy that AI can figure out proofs of just about anything we could think to ask, in milliseconds, for a penny. In such a case, there's really no nee
286.
by daxfohl 1y ago
Actually it would probably prioritize self preservation over energy conservation, so it'd at least continue maintenance and, presuming it's smart, continuously identify and guard itself against potential problems. But even that wi
287.
by daxfohl 1y ago
There still needs to be someone to ask the questions. And even if it can proactively ask its own questions and independently answer and report on them to parties it thinks will be interested, then cost comes into play. It's a finite re
288.
by daxfohl 1y ago
True, Many fish are (as far as we can tell from stress chemicals) perfectly happy in solitary aquariums just big enough to swim. So LLM may be perfectly "content" counting sheep up to a billion. Silly to anthropomorphize. Whatever
289.
by daxfohl 1y ago
Maybe give them some options to increase stimuli. A web search MCP, or a coding agent, or a solitaire/sudoku game interface, or another instance to converse with. See what it does just to relieve its own boredom.
290.
by daxfohl 1y ago
I wonder if the LLM could figure that out on its own. Maybe with a small MCP like GetCurrentTime, could it figure out it's on a constrained device? Or could it ask itself some logic problems and realize it can't solve them so it m
291.
by daxfohl 1y ago
Sounds like they could benefit from the R (language) build model that checks library changes against registered consumers of that library before merging. https://jtibs.substack.com/p/if-all-the-world-were-a-monorep...
292.
by daxfohl 1y ago
Right, which is why I asked.
293.
by daxfohl 1y ago
Oh, absolutely. The only real career path they have terminates at partner, which is basically sales. There is formally an option to stay technical, but people will look at you funny and you're not going to grow or get paid much compare
294.
by daxfohl 1y ago
I think there will be a couple benefits of using agents soon. Should result in a more consistent codebase, which will make patterns easier to see and work with, and also less reinventing the wheel. Also migrations should be way faster both
295.
by daxfohl 1y ago
Makes sense. Knowing nothing about it, I was picturing a tech sitting at home looking at pictures saying "yup, there's a spot", "nope, no spot here".
296.
by daxfohl 1y ago
My experience was 25 years ago, but back then we analysts and consultants were making around 50-60K and they were charging clients 5-15 times that when we were staffed. So I'm sure having a good buffer of people waiting and ready to go
297.
by daxfohl 1y ago
Are these the ones making 500K? Sounds like more of an assistance job than an MD.
298.
by daxfohl 1y ago
Yeah, I remember he commented on every town having its own local newspaper too, which has obviously been replaced by commercialized mass media today.
299.
by daxfohl 1y ago
And actually my bigger takeaway was to be very very careful when adopting higher-level tools. Things that may seem to match requirements on the surface could easily have the same kind of mismatches encountered above. A recent example of thi
300.
by daxfohl 1y ago
Don't do this. I did this many years ago for a small internal "parts" database for our small EE team, since we needed an audit history of changes. It was just awkward to use. Diffs were weird and hard to wrap a UI around, sea

More ›