Hacker Newsnew | past | comments | ask | show | jobs | submit | lijok's commentslogin

FUCK NO. Who in their right mind would let an LLM connect to prod?

Maybe at a greenfield startup. Where I work this idea wouldn't be entertained for a millisecond.

Many places have "dev", "test" "prod"... but IMHO you need "sandpit" as well.

From an ops point of view as orgs get big enough, dev wraps around to being prod-like... in the sense that it has the property that there's going to be a lot of annoyed people whose time you're wasting if you break things.

You can take the approach of having more guard rails and controls to stop people breaking things but personally I prefer the "sandpit" approach, where you have accounts / environments where anything goes. Like, if anyone is allowed to complain it's broken, it's not sandpit anymore. That makes them an ok place to let agents loose for "whole system" work.

I see tools like this as a sort of alternative / workaround.


Sandpit should be a personal (often local, if possible) dev environment. The reason people get mad about dev being broken for long periods of time is that they cannot use dev to test their changes if your code (that they depend on) is broken in dev for long periods of time.

Agreed on all points. Local loops are faster and safer wherever possible.

But particularly for devops / systems focused work, you lose too much "test fidelity" if you're not integrating against real services / cloud.


There’s no sandpit, only prod and dev, and you’re not allowed to break prod. Your developers work in partitions of prod. Dev is used for DR and other infra testing.

Well that’s just - dumb

Wanna elaborate?

Account vending machines where every dev can spin up thier own account is a thing and still under the control of some type of guardrails.

Hey, I get it. I don't want LLMs on prod at all. I made this to let agents connect to production cloned sandboxes, not production itself. I hope this helps your concerns, but I understand either way. Lmk with any other questions.

What’s a production cloned sandbox? Take my comment as feedback that the landing page is anaemic

For example, if you had an on-prem footprint with thousands of VMs, a production cloned sandbox would be a clone of a VM to let AI safely make changes, install packages, etc.

Yeah, working on the landing page. Feel free to ask any other questions!


why does it have to connect to prod in order to be useful?

I think you would be very surprised at a) how useful it would be and b) how lax prod can be depending on the company culture and stakes.

Classic prisoner's dilemma

how is there a prisoner's dilemma pattern of payoffs here?

Microsoft defected, now the correct move for the customer is to defect. Really, based on the pattern they learned, they can defect a thousand times before they lose enough customers over it that they have to publicly apologize and reverse a feature, so customers stepping in mid-game think that there's a tit for tat happening, but they don't realize they're starting off in the hole. We all should have left Microsoft a very long time ago.

Does Kessler syndrome also mean ICBMs become nonviable?


No.

It's not a wall. The risk from going through a dangerous orbit is much much less than the risk from staying there.


That depends on how you define risk. If it means the probability of a collision, then you'd be correct. But if a collision does happen, the consequences will be worse than being in the same orbit. Based on an oversimplified model, debris in orbit is likely to have low relative velocities with respect to an intact satellite in the same orbit, since a large deltav would change the orbit. (It's not as simple as this, but it's good enough in practice.)

This is actually what asat weapons take advantage of. They usually don't even reach orbital velocity, just like ballistic missiles (of course, there are exceptions like the golden dome monstrosity). The kill vehicle just maneuvers itself into the path of the satellite and lets the satellite plough into it at hypervelocity.


I remember a short story about Canada preventing total global annihilation in WWIII, by deliberately triggering Kessler syndrome. My google-fu is failing me though.


I would love to read it:)


> to the surprise of absolutely no-one with even the most basic grasp of how economies function

So roughly 98% of the population was surprised?


Like Python?


CUE


Why not Python?


Typing is bolted on rather than a native concept, for one.


Why is that a problem?


Because types are important and having them be a native part of the language creates opportunities for error checking, editor completions, and LLM bounding.


Invisible scoping and turning complete

Python is better than bash in ops, been using more Go in this space

Config is another beast and separate languages


I’m not sold that config is a complex enough domain to necessitate another language. What problems is CUE solving when compared to python and why are those problems substantial enough to make it worth learning a new language?


That's exactly the thing -- complexity. Cue bounds complexity, like json, yaml, and toml. But it offers more composeability than any of them.


Given that we now have TOML, JSON, INI, CSV, YAML, etc it seems we are converging on either JSON, YAML or TOML. There is too much inertia behind those three and not much behind CUE right now.


CUE works with all of those languages, so it doesn't matter what the tools or others are using. I can always apply CUE at any point to output their required format as needed.

Keep your legacy config and mess if you want, you're the one missing out

Also, I don't see TOML in the wild enough and the others have been around long enough, I must chuckle and not take seriously these claims about "inertia"


I’m not claiming inertia makes TOML ‘best’, just that it’s clearly not blocked by inertia either. Cargo standardized on TOML years ago, and GitLab Runner has relied on it for a long time. If a format can win in major ecosystems, “people won’t adopt anything new” isn’t the whole story.”


I guess by "compatible" you mean the data plane.

There are choices that speak the S3 data plane API (GetObject, ListBucket, etc).

There are no alternatives that support most of the AWS S3 functionality such as replication, event notifications.


None? I've seen a few projects that purport to be a drop-in replacement for S3.


Can GitHub change their API response rate? Can they increase it? If they do, they’ll break my code ‘cause it expects to receive responses at least after 1200ms. Any faster than that and I get race conditions. I selected the 1200ms number by measuring response rates.

No, you would call me a moron and tell me to go pound sand.

Weird systems were never supported to begin with.


Not surprising. Take any conference and look at the schedule of some CEO or other “socialite” attending said conference. They’re not in the building, they’re running around town attending meetings. At JPMHC everyone is a “socialite”


At no point in history has humanity ever cut back on spending after some constraint got alleviated. Exact opposite, we always ramp spending up to chase new possibilities.

If A.I maximalism gospel was true we would see companies raising absurd seed and A rounds in record numbers. Which is exactly what we’re seeing


How many lives would AI have to save for you to say the energy cost is worth it?


I see no point in making this a numbers game. (Like, I was supposed to say "five" or something?)

Let's make it more of a category thing: when AI shows itself responsible for a new category of life-saving technique, like a cure for cancer or Alzheimer's, then I'd have to reconsider.

(And even then, it will be balanced against rising sea levels, extinctions, and other energy use effects.)


> when AI shows itself responsible for a new category of life-saving technique, like a cure for cancer or Alzheimer's, then I'd have to reconsider.

We’re way past that


Please, go into detail!


Here’s a good exercise;

Search through github for commits authored by .edu, .ac.uk etc emails and spend a few days understanding what they’ve been building the past few years. Once you’ve picked your jaw off the floor, take another 10 minutes to appreciate that this is just the public code by some researchers, and is crumbs compared to what is being built right now behind closed doors.

Tenured professors are abdicating their teaching positions to work on startups. Commercial labs are pouring billions into tech that was unreachable just a few years ago. Academic labs are downscaling their interns 20x. Historically hermit companies are opening their doors to build partnerships in industry.

The scale of what is happening is difficult to comprehend.


AlphaFold?


How many lives have been saved by AI? How many lives have been lost because of it?


Not what I’m asking. But idk, do you have stats? I wouldn’t say _lost_ as a ding against, _ruined_ or _negatively impacted_ is sufficiently a problem


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: