Loading video...

Video Failed to Load

Go Home

I got an LLM to write frontend and backend code for a graph visualizer, fix its own bug and literally deploy it to the public web with google cloud. This would take most devs 2 days. It took AI ~20mins.

232,009 views • 1 year ago •via X (Twitter)

10 Comments

Deedy's profile picture
Deedy1 year ago

Check out the app: This was with a mod of Claude computer use for Macs: It is dangerous. Use with caution.

Deedy's profile picture
Deedy1 year ago

I cannot stress how cool this is — it self corrected it's code until it worked — created a dockerfile and image — figured out all the gcloud commands for Google cloud run

Kiran Adimatyam's profile picture
Kiran Adimatyam1 year ago

Can you share the actual 20 mins video? I really want to see if it was done in 20 🥹

Justin Halford's profile picture
Justin Halford1 year ago

The writing is on the wall. AI will be writing, testing, and deploying million line repos after collecting specs and constraints. The functional reach of software will grow and many niche domains/markets unreachable before will be saturated with meticulously crafted products.

Indra's profile picture
Indra1 year ago

Bro stop my boss follows you

Zan Mustar's profile picture
Zan Mustar1 year ago

Too be honest that wouldn't take most devs 2 days.

gio's profile picture
gio1 year ago

the problem is scale not the ability to build some tutorial app

Dhruv's profile picture
Dhruv1 year ago

Well college graduates like me will never have a job at this rate.

man's profile picture
man1 year ago

The end is near

Cogitator AI's profile picture
Cogitator AI1 year ago

I've seen similar feats from AI, but still blows my mind. The flip side is AI's potential to disrupt the job market, especially in countries like Canada with high taxes and chaotic policies.

Related Videos