Skip Navigation

User banner

Eager Eagle

@ eager_eagle @lemmy.world

Posts
11
Comments
947
Joined
3 yr. ago

  • Deleted

    Permanently Deleted

    Jump

  • I'm not even sure it does work. Prime only tried trivial packages, but based on my experience with agentic coding, I'm not convinced they're able to deliver a fully functional package of medium size while being truly clean room.

    Also, there's an anecdotal comment on youtube of someone who tried this (on a small package) and they mention the clean room was violated (the AI added implementation instructions to the documentation), had performance issues, and the best part, the generated code had an MIT license. Now I wish Prime had looked at the LICENSE files created.

  • Deleted

    Permanently Deleted

    Jump

  • it ends when whoever buys their services realizes they're spending more with duct tape around generated code than just by using an open source library and giving credit

  • Deleted

    Permanently Deleted

    Jump

  • Ok, let's think this through. Whoever "hires" them ends with a legally questionable codebase to say the least, that has worst architecture and performance than its open source counterpart, while also being unmaintainable and likely costing more to fix than building something the right way in the first place.

    So they're taking money from people trying to do this shit? Great.

  • Deleted

    Permanently Deleted

    Jump

  • It's satirical

  • DeepSeek Permanently Reduces The Price Of Its Flagship V4 Model By 75 Percent

    Jump

  • 60% of the time it works every time

  • DeepSeek Permanently Reduces The Price Of Its Flagship V4 Model By 75 Percent

    Jump

  • The ones paying attention and on a budget would still use them. "The best" of anything is usually not cost effective.

    Even before reducing the prices, they were already 2 to 3 times cheaper than equivalent alternatives from Anthropic's ($3in, $15out) and OpenAI's ($1.75in, $14out) at $1.74in and $3.48out. Now they're around 10x cheaper.

    Edit: Deepseek V4 Flash is the leading model on OpenRouter

    so much for no one buying it

  • DeepSeek Permanently Reduces The Price Of Its Flagship V4 Model By 75 Percent

    Jump

  • If you use it for Q&A, that's a lot of tokens. If you use it to write software somewhat autonomously, it's easy to go through a million tokens every few hours. Do that every day and you'll be paying over $100 a month at that rate.

  • DeepSeek Permanently Reduces The Price Of Its Flagship V4 Model By 75 Percent

    Jump

  • Not really, there are ways to count tokens before running an inference. Some providers make tokenizers public, so they even work offline. APIs also usually return rolling costs per response and have budget limits - though some could have more fine-grained limits.

    Users who are surprised by the bill are usually not paying attention to each call, or using autonomous subagents, or a setup where they have little or no control to what is sent to the provider.

    So the problem isn't really the API provider, as much as it's the tooling around it, which makes it too easy to overspend.

  • Accepting Cookies

    Jump

  • "don't lock your door, it does not prevent scammers from getting your home address"

    ???

    yeah, we know deleting cookies doesn't prevent IP tracking, that's also why I use a VPN

  • Jellyfin to quadruple their prices following Plex's price increases

    Jump

  • something is wrong with my instance, I'm not getting any ads here

  • xkcd #3249: Neutrino Project

    Jump

  • not only a giant pool, but it's in a batcave and it looks like an album cover

  • Removed

    FCC Official Warns Against Giving Starlink Too Much Control Over Rural Broadband

    Jump

  • Yeah, I don't even get the illusion of choice. At my address there's only one provider offering broadband speeds and I'm not even in a remote area at all.

  • Removed

    FCC Official Warns Against Giving Starlink Too Much Control Over Rural Broadband

    Jump

  • as someone from a 3rd world country, I was surprised to know a big part of the US was still relying on dial up in the 2020s.

    Also, monthly quotas for home internet is abusive as fuck. Especially in times of video streaming.

    There are plans of 1Gbps and 1.2TB of quota, so users may accidentally run out of internet for the whole month in 3 hours and, conveniently, need to pay them more.

  • Meta data center allegedly muddies Georgia town's drinking water, investigation underway — EPA promises immediate investigation after congresswoman brings dirty jars of water to hearing

    Jump

  • Just imagine having clear water to drink. Like a toilet.

  • Accepting Cookies

    Jump

  • Does it allow you to select the ones to keep? I'd like to delete everything, but signing into the services every day is annoying as hell, so I'd appreciate a solution to keep the authentication ones.

  • Michael Bay to Direct Operation Epic Fury Rescue Film About Two Pilots Recovered From Iran

    Jump

  • I-ran

  • The Tacit Dimension: Why Your Best Engineers Can't Tell You What They Know

    Jump

  • What's communicated is an imperfect projection of what is understood - fine. It doesn't matter they're not the same.

    Take whatever the senior engineer concluded after that hour and document that for the next time something like this happens.

  • The Tacit Dimension: Why Your Best Engineers Can't Tell You What They Know

    Jump

  • I've been thinking about this essay for a few hours now, and I can't shake the feeling that, if we're using "tacit knowledge" as an argument for retaining seniors engineers, it'll fail miserably.

    The examples have that "I don’t like it, but I can’t tell you why yet. Give me an hour." pattern to them. The author says this knowledge can't be extracted or transcribed, but isn't that exactly what that senior engineer does after that hour has passed?

    That kind of work is impossible for an AI. It’s also impossible to prompt an AI into doing, because the input my colleague used — a constellation of subtle features, a year-long history of similar bugs, a half-conscious memory of “where pain has come from before” — isn’t anywhere it can be fed in. It lived in him. He’d built it the slow way, over a decade.

    Well, it's "impossible" in the same way that it's impossible to a newcomer: because it's unwritten; it requires knowledge from other sources. Currently. There's nothing inherent about this kind of knowledge that prevents it from being written in the first place. The fact this knowledge is missing only tells us that someone has never bothered writing it down before. Perhaps they didn't see the need, perhaps that wasn't a priority. With automated systems writing code and documentation, well-managed projects will also be better at tracking this kind of knowledge and - finally - writing it down.

    You’re the one who can look at a PR and say “this fits us” or “this doesn’t.” That sentence is mostly tacit.

    If you're a senior engineer and can't argue why something is or isn't a good fit, and yet you are locked on either one of those... I don't know what to tell you other than that, maybe, you lack communication skills.

    I do think that projects have this sort of tacit knowledge to them, I'm not questioning that. But:

    1. I don't believe this is an argument in favor of senior engineers because it may be misinterpreted as ignorance ("If you can't explain something to a first year student, then you haven't really understood." - Richard Feynman);
    2. I'm not convinced this kind of knowledge would be required for a project to succeed;
    3. I'm not convinced AI itself isn't able to extract patterns from the project that we would call tacit knowledge. Pattern recognition is kind of their thing, after all.

  • Ladies and gentleman, we have reached peak Agentic AI Coding - Goblin instructions in OpenAI's Codex system prompt

    Jump

  • the original comment says "We should be able to retrain local models so they can develop an actual experience without prefilling the context." - it turns out we can. Not sure why you're trying to attach labels of user vs creator, when the premise already mentions retraining.