Hacker News new | past | comments | ask | show | jobs | submit | risyachka's comments login

The price is set by the market. It never was and never will relate to the seats/resources used/etc.

The price is set by the market as a function of some sellers charging by seats, others by resources used, etc, and some buyers preferring simple pricing models, others preferring usage-based, etc.

Looks like it. The only other explanation is they just stopped investing in other products and focused on Agents, thus no need for loads more of engineering.

No way in the world they got 30% gains. 5 maybe realistically


I recently had a refreshing conversation with a bank VP (head of AI), he said yes, they do see 30-40% improvement in "some" processes, so overall maybe 0.5-1% improvement.

So I'm betting they got 30% gains in e.g. "OCRing" and quote that, ignoring that the OCR part is 1% of an entire process chain.


Does OCR refer here to “optical character recognition”?


Yes. Just using that as an example where "AI" (not the hyped up LLMs) has really helped.


Their ultimate agent product strategy feels like reducing cost of sales... by automating salespeople.

I have doubts that will ever happen, but who knows?

Sales folks are highly-compensated, so even making a few of them redundant (or making existing salespeople more efficient) would be a big win for companies (Salesforce's customers).


I think a lot of CEO were sales people.

Being in that field, they can see where LLMs would fail at it.

They can understand that you can’t just replace a sales team with LLMs.

They don’t have that for engineering…


Im an engineer. I see no reason why LLMs shouldn't already today replace most people in sales /s


Compensated proportional to the revenue they bring in close to the time they bring it in. (Operations cost)

Very different from the software dev who gets paid today for revenue that might come in in two years. (Capital cost)

The average person has no idea of how capital sees the world. A worker feels resentful if they get paid less per hour of input, a capitalist feels resentful when they get paid less per dollar of investment. The Marxist viewpoint that they conflict directly is quite wrong: operations costs can be passed on to the consumer, but a capitalist is going to have to negotiate with their investors if they are having trouble with the bang/buck ratio of their investments.

(I'd had a job go really badly. A friend of mine said my problem was "I was only getting paid a fraction of the value that I create", I said "I tried getting paid more than the value I created and it ended in tears")


Why do people think costs can get passed on? They can't be. Pricing is about supply elasticity and demand elasticity. Sometimes you just need to eat costs because of the competition.


Often though the competition is facing the same pressure. Or there is no competition.

My trainer at the gym introduced me to

https://patents.google.com/patent/US10245460B2/en?oq=1024546...

which sells for about $250 (worth it if you really use it.) My first instinct is that this product ought to be available for $25 on Temu if there wasn't a patent but I know from experience that if I talked to folks at TRX they'd have a good explanation of why my number is low. (e.g. TRX is in a position to pass costs on)


I got one of those from walmart in like 2018 for like $20, still going strong


TRX won a lawsuit and you can't get one from Wal-Mart today

https://www.prnewswire.com/news-releases/trx-obtains-federal...


These are between $20 and $250, but having many different band resistance levels is useful:

https://gofit.net/collections/resistance-tubes-bands

There are plenty of cheaper no-name options on Amazon, but they have the property that the handles snap off during use, risking injury.


TRX is a rigid band that doesn't stretch. You set the "resistance" by the angle of gravity relative to the band.

A TRX push-up puts less load on the primary path than a conventional push up but is very challenging to all the other muscles that it takes to not flop over when you do it

https://www.youtube.com/watch?v=DNuZWO0if5o

These days I'd trust AMZN less than Temu. At least Temu and sellers on Temu still want to win your trust. AMZN thinks they have it and will still think they have it long after it's lost.


This is less about Marxist/capitalist and more about a change in regulation in the US that makes R&D expenses (dev headcount) amortized over five years (for taxes) versus sales being a straight-up expense. This one is going to be good for the stock price, and the AI Agent angle is a marketing masterpiece.


It's a good point about changes in tax laws for software devs. Makes me glad I work for a non-profit (a rare non-pathological non-profit no less!)

For all the discussion about a soft job market for devs, the fact that the last Trump administration made a tax change that puts a target on our backs comes up rarely but it is part of the explanation.


The chance happened in 2023.

The investors are the capitalists. The C suite negotiator is petit bourgeoisie.


Customers will install agent interaction prevention agents to prevent the infinite onslaught.

Sales will still happen with dinner and handshakes.


Salespeople are typically compensated based on commissions. At least the well compensated ones are. Automation can make things easier and streamline the sales process, but because the sales rep is paid a percentage of the GP of a sale, automation doesn't really their take.


The company still pays that commission. If the company instead kept that money, it would be more profitable (or could lower the price of its product).

We can get into the nuances of advantageous tax treatment or not, but sales commissions come from somewhere and are essentially transaction costs.


There's probably space for AI secretaries to screen contacts and to drop AI sales.


Indeed, it's marketing slop as a pretense for a hiring freeze.


This kind of smokescreen only works for so long. Once it all starts failing and profits dip, things will change.


>>“Better product”: We need to define "better" clearly, but if you're basing this off your R&D efforts, I would very much fear the competition coming my way.

Yeah, no, better product will always be a strong moat.

Competition could copy bette products for decades now without ai, but most software today is trash.

they always "could", they never "will"


>> The AI went from: >> "This file contains authentication logic using JWT tokens"

So what was the initial prompt? "What's in this file?"

And then you added context and it became context-aware. A bit of an overstatement to call this "Holy Shit moment"

Also , why is "we"? What is "our AI"? And what is "our benchmark script"?

And how big is your codebase? 50k files? 20 files?

This post has very very little value without a ton of details, looks like nowadays everything "ai" labeled gets to the front page.


> looks like nowadays everything "ai" labeled gets to the front page.

it’s been this way for like a year or more. hype machine gotta hype.


It’s ironic that most people have jobs in the first place thanks to ads but hate on ads every day.

Without ads the world would basically stop, they drive the businesses, each and every one.

It’s a necessary evil that will always be in most popular products, especially considering most people want to get all value for free.


I want the power of the techno gods for free, but I also don't want ads.

Is that so hard to get?


Nope

It's like credit cards. It's amazing benefits wouldn't exist without leagues of people chronically keeping a balance against near-usury levels of interest in some cases, but, fortunately, the set of people that use them correctly is much much smaller.

Most people irl don't give a shit about ads.

Evidence A: NFL and soccer growing in popularity every year despite 90 sec of ads every 5-10 minutes.

Evidence B: Netflix cheap and Spotify free being more than enough for millions of people.

Without them, our entire industry wouldn't exist. Being able to dunk on ads and invest time on ad blockers is a privilege, really.


Netflix is interesting. The current plans are:

  $ 6.99 Standard with ads
  $15.46 Standard
  $22.99 Premium
The standard plans are 1080p. Premium is 4k UHF + HDR.

The differences between the standard plan and the standard plan with ads, aside from the ads, is that the plan with ads also: (1) only has "most" of the movies and TV shows Netflix, (2) and does not allow purchasing an "extra member" slot.

The difference between the standard and premium plans, besides 4K, is: (1) premium lets you download on 6 devices at once (both standard plans are limited to 2), (2) premium includes spatial audio, and (3) premium lets you purchase up to 2 "extra member" slots.

I'm really curious how many people who do not get Premium go for Standard rather than Standard with ads.

I subscribed a couple days ago, specifically to watch a show that was prematurely cancelled by its original broadcast network (on a cliffhanger!) that Netflix picked up and finished. I figured that since I had watched the first 3 seasons on broadcast TV with ads and don't recall being overly annoyed I could put up with ads for the rest, and I could always switch to a no ads plan if the ads were a problem.

So far, after 5 episodes of just under an hour each...the ads were barely noticeable.

It has shown me 3 ads, each 30 seconds long. They came at points where the was a major scene change so they didn't disrupt the show. They have, so far, been unobtrusive enough that I wouldn't even pay $1 extra to get rid of them, let alone pay more than double.


My understanding is that American football and other American games are built around tv adverts

How does normal football allow for tv adverts? (Other than half time)


it doesn’t :)


Ads haven't always existed. Economies without them can exist and we can move toward one.


Can you explain how a service like YouTube could exist any other way?

> Economies without them can exist

We should be able to both hate ads and understand that they are a necessary monetization model for certain things. If you feel strongly that they aren't necessary what model should be used instead?


Maybe YouTube can't exist without ads but the next thing might.


You can literally pay YouTube to avoid ads, watching the mental gymnastics folks on here do to complain about YouTube Premium is my favorite pastime.


Why would you pay for something that treats you as a product? I would gladly pay for youtube if there was: no paid sponsorships, no ads, no algorythm manipulations, no collection of behavioral information, no clickbaits, no shitty content promoted and so on and so. Basically a service where the side paying is a consumer, not a product. Paying for service that is already extracting value from you in ways that are not respectable is insane.


A well pruned YouTube subscription is right now the highest quality video entertainment and information service in the history of the world. You literally just need to "like and subscribe" to a bunch of high quality videos and YouTube will recommend you a never ending supply of high quality videos. You can then continue to improve your recommendations with the like/dislike buttons.

If you're seeing "shitty content", clickbait and paid sponsorship, it means you haven't told the YouTube algorithm what you want to see. It's just the push of a button.


> no paid sponsorships, no ads, no algorythm manipulations, no collection of behavioral information, no clickbaits, no shitty content promoted

There are plenty of channels that do none of these. As for "algorythm manipulations", YT recommends more of what you like or subscribed to. Not sure what you expect a content aggregator and provider to do. Just have a search bar? I've come across so many interesting channels thanks to the "algorythm manipulations".


I pay for YouTube premium. But it simply would not exist to the same scale or utility if that was the only model that was funding it. Incredible to see the mental gymnastics needed to avoid that fact.


> But it simply would not exist to the same scale or utility if that was the only model that was funding it.

Citation needed. Are you saying that if there were a billion YT premium subscribers, YT would still need ads for its scale? I doubt it.


I almost fail to see what harm could come from nationalizing YouTube at this point.


what adverts does YouTube have?


They have existed since humans could write or talk.


Market may be saturated, but 99.9% of all software and apps are unusable garbage.

So there's a ton of opportunity here.


I'm not sure about them being garbage but I do believe very strongly that existing SaaS software is way way wayyyy overpriced. So there is an opportunity to build a simple, low price software that does less but just works.


This is where I like to operate. My career is implementing ERP software. It's expensive and I notice clients really value about 10% of what it can do. So I built an opinionated version of that, focusing on being the best at that 10% and keeping the features limited to just that versus the ERP approach which is a platform for all but a solution to none.


Good one. I love this approach. Work somewhere. Find which part of the big software is most valuable and you can make a business out of it and run a small business.


Garbage as they mostly have too many features. You can do product with 30% of features and charge half their price.

Maybe you will end up getting like 25% of their customers but if you play it well that is going to be worth still a lot.


Excellent point. That's exactly what I'm betting on with https://neeto.com products.


SaaS software is cheap already. Try doing those tasks with pen and paper, or with inadequate tools.

If you're competing on price, you're usually competing for the worst customers. Good for them, but not so nice for you.


>If you're competing on price, you're usually competing for the worst customers.

This is true. Also, if you are only competing on price, it is easy for a competitor to undercut you.


Overpriced complaints always sets off alarm bells to me if it solves some real problem/people like it/etc. There are usually migration costs and "way overpriced" may still be a small slice of overall budgets.


True. For some companies that price is not enough to move the needle. For some companies it will be. Calendly is worth 4B at this time. If I could move 1% of their customers then that's worth it for me. Executing this won't be easy by there are way more than 1% of the people who will move to save cost.


> Market may be saturated, but 99.9% of all software and apps are unusable garbage.

in terms of marketing the question is how you reach customers, and demonstrate to them that your software is not garbage unlike other within 5 sec of their attention span.


Honestly if you are struggling to work on your side projects you are most probably doing just fine in your life and you don't really want/need them.

In this case why force it?


Isn't this a recipe to complacency? Things that are worth achieving take hard work. If I only do things that I _really_ want to do right now, I'm not sure I will live a fulfilling life. Especially when many of us have been trained by the internet to seek immediate reward above all else.


It has nothing to do with motivation.

They have $1 salary and then get a ton of dividends or sell stock (both of which are taxed at waay lower rate than salary)

So they are very well compensated at all times.

I honestly don’t understand why they like to mention this $1 bs all the time.


True, but at the same time immigrants have nothing to lose and everything to gain, thus in many cases they will work harder/longer and as result often be more skilled than an average developer.

And in general skilled immigration has many times over been proven to only benefit the country and that java developer you mention.


This.

When this moment becomes reality - the world economy will change a lot, all jobs and markets will shift.

And there won’t be any way to future proof your skills and they all will be irrelevant.

Right now many like to say “learn how to work with ai, it will be valuable “ . No, it won’t. Because even now it is absolutely easy to work with it, any developer can pick up ai in a week, and it will become easier and easier.

A better time spent is developing evergreen skills.


Consider applying for YC's Spring batch! Applications are open till Feb 11.

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

Search: