29 comments

[ 2.2 ms ] story [ 48.9 ms ] thread
"NFTs show relatively small groups of tens of thousands of people can work together collectively for a common good" - Garry Tan, 23 March 2022

https://x.com/garrytan/status/1506769562468958210

"If you own an NFT..." you are an innovator among innovators. - Garry Tan, 30 September 2021 [quote tweet]

https://x.com/garrytan/status/1443460589049704450

Web3 vs Earlier incarnations technology adoption curve [image tweet]

https://x.com/garrytan/status/1521530531568963584

"airdrops going mainstream could truly upend some % of the centralized ad-based economy of Web2" - Garry Tan, 25 Dec 2021

https://x.com/garrytan/status/1474826162408808448

(comment deleted)
Will they lose a small customer or two to custom vibe coded solutions, yes. Will they be "competed away" due to vibe coding, definitely not.

You have to remember that no one gets fired for selecting an established software provider. Zoho is a 'safe' pick for SMBs because you know it will work compared to some custom CRM that Bill from accounting vibe coded in an afternoon.

He will be right at some point. The question is when.

I'm already vibe coding complex things like GUIs, my desktop environment (NixOS), and last week a Wayland layer shell client that would have taken me quite a lot of work if I had to do them myself from scratch from docs, and I have 20 years of software experience.

The things I spend time building and polishing today with my own time are just next year's vibe-coded minutia.

Some people are going to have a very hard time swallowing this pill, though.

Considering how much money they invested in countless AI VS Code forks and other vibe coding platforms, he needs to hype up the market.
> Why pay $30/seat/month for over bundled SaaS when soon even nontech ops ppl can vibe-code a custom solution in a weekend?

I don't get his suggestion. I'm paying Zoho for email hosting. What am I even supposed to vibe-code successfully for me to drop Zoho? There's no shortage of open-source IMAP/SMTP servers, in fact Zoho is probably using them too. Sure I can spin one myself, but I'm paying for Zoho for the *service*, not the software.

I kinda hope so. When I started my career companies and orgs had in-house developers to make custom software for the org. The developers understood the business and were part of the culture. Then the labor market got tight, developers left for higher paying jobs, managers became afraid of going custom because they would lose the developers, so they starting buying overpriced SAAS products and contractors. If the job market is flush again with capable devs armed with AI, maybe we can go back to in-house developers again.
Obviously not? Accounting software has to work or the IRS (or your local equivalent) will come after you. Zoho "just works".
> competed away by people building their own custom software built by people using @Replit @emergentlabs and @Taskade

Why do you need 3 platforms to vibe code a solution? Are each one of them not good enough?

Vibe coding can only affect companies that depend on slow-coding.
Well, then I claim that it will be the companies themselves that try to replace Zoho (or any other CRM for that matter) with a vibe-coded replacement, the ones going out of business. As someone said once: focus on what makes your beer taste better.
It's 'Garry' and besides that guy gets too much airtime already.
(comment deleted)
Let's say I'm a manufacturer of widgets, and instead of buying an off-the-shelf CRM I decide to have the resident IT whiz on the team vibe-code a custom solution for our needs. Now I'm a manufacturer of widgets AND a CRM SaaS shop, responsible for software maintenance/deployment/reliability/feature roadmap/bug remediation. I guess the idea is that AI agents will take care of all of those things too - to which I guess my perspective is "good luck! I hope that works out."
Most probably all VCs will go out of business as the cost of creating software companies approaches zero. The need to create an army of software developers is no longer needed.
Pichai should be fearing for his position after all the misallocation of capital and the low quality of Google "AI".

Of course he endorses slop coding (Pichai's endorsement is criticized by the Zoho founder for those who do not click through to X).

Zoho will now be a very interesting company for the vast majority of people who hate "AI". I'll have to check it out.

He said `would be first`; I 100% agree. If vibe coding was going to take over the world, then Zoho going out of business is a very good leading indicator. As soon as that happens, I will invest heavily; till then it is safe to ignore it.
> Why pay $30/seat/month for over bundled SaaS when soon even nontech ops ppl can vibe-code a custom solution in a weekend?

Feels very disingenuous. I'm a huge proponent of AI drastically increasing efficiency of creating software but we're a long ways away from nontech people replacing and supporting collaboration tools used by medium sized businesses.

> ...when soon even nontech ops ppl can vibe-code a custom solution in a weekend.

These statements are so out of touch with reality, I generally wonder where will YC be in 5-10 years.

Technology alone rarely wins a market ... success usually comes from marketing, referrals, and network effects.

It’s the same reason why vibe coding a better version of Airbnb (even if it’s just a simple CRUD app) wouldn’t actually threaten Airbnb as a business. The product isn’t the moat; the ecosystem is.

This is the guy that said NFTs were the next "big thing"?
> competed away by people building their own custom software built by people using @Replit @emergentlabs and @Taskade

Using same logic, why use Replit, EmergentLabs, or Taskade when you can vibe code your own vibe code platform?

> Why pay $30/seat/month for over bundled SaaS when soon even nontech ops ppl can vibe-code a custom solution in a weekend?

Seems ironic to post that - doesn't that same logic imply that most YC companies are worthless?

This going from the front page to effectively buried (cant even find it in the first 10 pages) within minutes is not a great look for HN.