Video wird geladen...
Video konnte nicht geladen werden
Google Cloud AI engineer just showed how they go from idea to deployed app at Google in 30-minutes using Claude. 26-minutes. free. by Google AI team. one person + Claude + Google Cloud = a full engineering org running on a laptop. worth more than any $500 vibe-coding course.
1,781,250 Aufrufe • vor 4 Monaten •via X (Twitter)
43 Kommentare

No one gonna use the app. TL;DR introduce the google MCP and kit for agent to do google cloud task. the rest of video is about make simple wireframe and make it look better - from engineer perspective this is useless but this can wow non tech for sure

Thnx, gonna watch it before bed

you are welcome brother, have a great watch !

He looks so much like a professional hahaha

he is very professional bro, imagine Google team shipping with Claude, crazy

TL;DW No, he didn't build anything in 30 minutes. He was talking for 30 minutes about AI.

26 minutes to ship. 26 hours to understand what you shipped. The bottleneck didn't disappear — it moved.

If that is the case, and anyone can do it. Google apps no longer have any values. And if by chance yours does, it will never be found in sea of other apps.

“The real disruption isn’t AI replacing engineers it’s one engineer suddenly shipping like a 20-person team.”

All @polymarket trdading bot developers. Come here, and watch @0xMovez 's post, He is posting really pricessless content, and helpful for the development. i also developed many bots based on his post/articles. Thanks , my mate, @0xMovez !

@Polymarket Thanks, buddy, appreciate your support.

Is this on YouTube?

The biggest shift is that cloud + AI turned distribution for building into leverage. One person can prototype, ship, and learn faster than small teams used to.

thanks for support mate, amazing watch, booking will watch today

thank for your share too mate

Time to watch

real alpha here brother

Your posts are always alpha

The real shift isn’t just “vibe coding.” It’s compressing idea →prototype → deployment into hours, not months. As AI handles scaffolding, testing, infra wiring, and docs, teams get smaller but more leveraged The bottleneck becomes judgment, architecture, and product thinking.

Damn, dude, thanks, it's literally free money, it just takes a little time to figure it out

you are welcome buddy, happy its useful for you

let me watching this video my bro like learning something new in AI

Why they don’t use Gemini? It is Google… or maybe I don’t understand smth?

wait... but y claude? y not their models?

one person + Claude + Google Cloud = a full engineering org running on a laptop This is the equation I've been running at 18. Built and shipped real products solo. The "team" is now just you and the model.

@0xMovez wild. never thought I'd see the day when deploying apps was faster than my coffee order 😂

The real unlock is not “build app in 30 min”. It’s removing the waiting time between roles. No handoff from PM to designer to backend to DevOps to QA. One person can now move through the whole stack in a single loop. That loop speed is what kills old software economics 🚀🚀

Like it. The $500 course taught you the theory. the Google engineer just made it free and live

This is why the learn to code or die takes are aging badly. One person with taste + product sense + AI tools can compress an early-stage team into an afternoon now.

One person + laptop + Claude replacing a full eng org is not a drill anymore 🔥

great share my man

another banger share, watch it

People need to stop saying “free” with respect to using Claude or ChatGPT to do large tasks. These services are both extremely expensive currently.

this buddy say very smarts words

I have to spend 2 weeks and find 12 people from Android who have to confirm that my app (which has been on iOS for half a year) can be published in the very respected Play Market. And I will be very lucky if I manage to pass this "test" for the first time.

Then why do they still have 80000 engineers at Google and still hiring for that matter? Something doesn’t add up.

@harshexists Why did they not use Gemini?

This is the real shift. The stack matters less than the feedback loop now - one person who can prompt, judge, and ship fast can outpace teams still moving like committees.

I actually just did this for two projects today and it was wonderful! Just had to enable billing and create the database in the project via UI. I'm thinking Google's existing infrastructure is perfect for agents.

Thats the real unlock. The tool matters less than compressing idea deployed feedback into one sitting. Once you can do that in 30 minutes, shipping beats planning.

love the demo but this needs a disclaimer: he's a google ai engineer. when claude quietly breaks security or racks up a bill, he catches it in real time. non-technical founders get the 26-min deploy. they don't get the safety net. vibe coding here ≠ non-technical people.

The 30-minute demo is the easy part. Production needs recovery logic, monitoring, and constraint files that keep agents from drifting. That deployed app becomes 6 months of hardening before anyone trusts it with real traffic. Real engineering starts after the deploy button.

the 30-min demo always works. weeks 2-6 are auth, state, and the form that breaks on safari, which is where most of those 'full app in 30 min' builds actually live or die. we built mk0r to harden that week-2-6 layer, auth flows and safari form quirks shipped as drop-in modules, written with ai
