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

At $enterprise, we were just looking for a proper term that sets "responsible vibing" apart from "YOLO vibe coding". We landed on "agent assisted coding".

It's a bit more technical. And it has a three-letter acronym. Gotta have a three letter acronym.


I like "YOLO vibe coding" or maybe "YOLO vibing" for short, if the context is clear :-)

Hmm another idea is "extreme vibe coding" as opposed to "extreme programming",

but those who did "extreme vibe coding" wouldn't know what it meant


AAC / Agent Assisted Coding is a good term.

> AAC / Agent Assisted Coding is a good term.

Yes, please don't push "vibe engineering" to mean how you defined it in your blog post. To me, it means exactly the opposite.

I see "vibe" as pejorative. Adding "engineering" does not elevate it from "vibe coding", as I think is your intention in the post, it just shifts "vibe" term to a different domain.

To me, "vibe engineering" means using LLM to develop "design" with no care as to its validity just like "vibe coding" means for "code".

"Agentic xyz" or "Agent assisted xyz" is more fitting.

FWIW, I do not see "vibe" as always pejorative, rather it depends on goals. When quick results and not long term quality matter, "vibing" is a legit tactic.

Anyways, just my interpretations. Please, keep up the good work. Remember, the two hardest things in software are naming, cache invalidation and off-by-one errors. It's good you continue to tackle the zeroth one.


A2C?

I really like "agent assisted coding". I think the word "vibe" is gonna always swing in a yolo direction, so having different words is helpful for differentiating fundamentally different applications of the same agentic coding tools.

abbreviation ass.coding.

Nice! I love how mermaid keeps getting better!

For sequence-diagrams, nothing beats https://sequencediagram.org/ (I am not connected with them in any way, just a happy user)


I've used this in the past for collaborative diagramming sessions and love its ease and simplicity, but the point of Mermaid is its portability - ie. can be embedded in Markdown docs and viewed in various editors/platforms.


Thanks @maho! We're hoping to keep the improvements flowing. I'm non-technical but from my perspective I thought Mermaid sequence diagram functionality really shines! Would love to fill the gap in my knowledge. What is better about https://sequencediagram.org/ than Mermaid sequence diagrams?


Agreed. I’ve tried other editors but so far I’ve always come back to sequencediagram.org — it just works, low fuss.


FWIW, I find MermaidChart's sequence diagram editor better than the sequencediagram.org one!


sequencediagram.org is using PlantUML I believe, so you can do these locally without their website.


I need this, but for heatpump noise.

It's mostly broadband noise that can be simulated by simpler methods, but visualizing possible resonance patterns for the low-frequency emissions from the compressor (which typically runs at 20Hz, 40Hz, ..., 120 Hz) would be good to know.

Although I am not sure how the 2d simulation result carries over to the 3d world...


The pronounciation of math symbols is hilarious, but not super useful. Prompt: "Give me Maxwell's equations".


I really like nools, which is a drools clone, but for JavaScript. It's fantastic for quick hacks and for getting to know how to write code for rule engines.

Sadly, it is no longer maintained.


Let's say I want to have a somewhat genereric CRUD-app: A database, some application logic, a login system, logging/telemetry and a UI.

What kind of stack should I use so that tools like Darwin (or just plain old copilot) can be most effective?


Most effective in what way? Are you looking to generate an app from scratch or have AIs work on an existing project?


From scratch


While Darwin can generate code for you, I think generating new projects from scratch is already being done by a lot of the major frameworks. Check out the native docs/tooling around the kind of stack you'd like to build!


My general answer: any tool that has a large online community. I’m not the dev, but any LLM-backed responses will naturally depend on that LLM’s familiarity with the tech. With that in mind, RubyOnRails and create-react-app (+ some node backend) seem like the natural winners.


This is insane. The project includes the hardware (GHz-capable RF-generation and measurement), firmware (FPGA) and Software (a cpp GUI). Surely that can't be all from one person?


It's under development close to 4 years. If the person has time (or doing Ph.D. or something on the subject), this is a very viable time frame to do all of this.

There are other people's work in it, but it's a one man show mainly.

This is a passion project and it shows.

Edit: Plus, this is the second iteration of the hardware. First iteration is at https://github.com/jankae/VNA


It's not that hard (not that it's easy either!). The individual parts are conceptually relatively simple, the devil is all in the details. For a generalist this is a doable, but likely very time consuming project. I've done something similar (fairly different focus in purpose and specs, but the overall shape and scope is not much off) professionally, mostly on my own.


"Surely that can't be all from one person?"

It's possible. Especially if they aren't a piece of shit like me.


Are we the same person?


I guess that's the existential question - is one pos really any different from anotherin a meaningful way; is any person in general?


Is there a way I can give feedback on wrong labels? The easy questions seem to be correct most (all?) of the time, but I noticed a few errors in the labelling of the complex question/answers. I would love to see this improve even further!


Any feedback like this helps -- shoot me an email at max@talc.ai with the name of the topic you saw incorrect labels on.

We didn't expect this much traction on the demo, or I'd have built this functionality in!


If memory serves correctly, Aldi Nord was one of the last big supermarket chains in Germany to introduce scanners at the registers (2003?), because their existing system was simply faster: Each item had a three-digit code, and all cashiers knew all codes by heart.

It was a race between me placing items on the conveyor belt and them ringing the items up. Oh the embarrassment when they told me the total as I was placing the last item on the conveyor belt.


> Aldi Nord was one of the last big supermarket chains in Germany to introduce scanners at the registers (2003?)

Aldi US is owned by Aldi Süd. Aldi Nord Owns Trader Joe's.[0]

They are two distinct entities now which carve out the markets between them.

[0] https://en.wikipedia.org/wiki/Aldi


After skimming the second paper, I still don't understand how precision mass measurements come into play here. They mention Cavendish-type measurements, but they are used for measuring the gravitational constant. Of course, you can turn the formula around, plug an unknown mass into the apparatus and then call it a mass measurement, but it's going to be a very imprecise measurement. A Penning trap can give you 11 to 12 significant digits -- a Canvendish-type measurement could give you maybe 5 or so, I think.

Or is it because the Penning trap measures "inertial mass" but they really want a measurement of "gravitational mass"? But wouldn't inertial mass fluctuate the same way?


(I went to a talk by Oppenheim author a couple weeks ago on this topic.) The idea is that gravity, as a force, only operates classically. More precisely: there is a classical state describing the curvature of space time, and then a quantum state describing the configuration of particles on that spacetime. But then, that quantum state needs to affect the classical state again (mass bends space), which would usually lead to the classical half becoming quantum and entangled with the other half.

You can keep the classical half (the shape of spacetime) classical, if the effect of the quantum part is partially stochastic. There's a minimum amount of random noise you need for it to be mathematically consistent. So, you set up an experiment where a particle is acting on another via gravity. There's a quantity of noise you should expect to see in the gravitational force.

"Inertial Mass=Gravitational Mass" now only holds on average. The gravitational mass will effectively have a Brownian noise term added in.


If this hypothesis is true, would it give us a way to distinguish many-worlds vs. wave function collapse? If the many "worlds" are all interacting with a single classical spacetime, we should be able to measure the gravity of other worlds, right? I'm not a physicist, but that sounds a lot like dark matter to me.


There is no implication that the classical spacetime wouldn’t still split into branches, possibly with different variations in the stochastic effect from each other.


Answering my own question based on Oppenheim's lecture[1]: In his theory, the stochastic interaction between quantum and classical states necessarily causes the wave function to collapse, so it is not compatible with many-worlds. Being able to measure the gravity of other quantum worlds is something predicted by semiclassical gravity, which Oppenheim calls "complete nonsense".

[1] https://www.youtube.com/watch?v=sde7k3jJp5E


So if I am understanding this correctly:

Quantum particles can effect change (curve) spacetime without direct quantum action if you sprinkle a bit of randomness into the (quantum acting on spacetime) effect?


Wouldn't it be enough to measure the (fluctuation of the) total gravitation force ( = gravitational constant times mass) exerted on the second particle, in order to draw conclusions about the nature of the gravitational force at small scales?


Consider applying for YC's Winter 2026 batch! Applications are open till Nov 10

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

Search: