Loading video...

Video Failed to Load

Go Home

I could not believe it's that easy to build an AI-powered recommendation system 🤯 In one hour, I built the ChatGPT for Movie Recommendations 🤖 You ask it a question; it recommends movies 🍿 Wanna learn more?

27,307 views • 3 years ago •via X (Twitter)

11 Comments

Daniel Henrique's profile picture
Daniel Henrique3 years ago

Using Supabase, interesting.

IMRΛN's profile picture
IMRΛN3 years ago

pg_vector and Supabase RPC ?

imVKartik's profile picture
imVKartik3 years ago

yep , very impressive indeed

Vadim Savin @notJust.dev's profile picture
Vadim Savin @notJust.dev3 years ago

And super easy

farz's profile picture
farz3 years ago

Best horror movie Scary movie 2 😂

Thato Lebethe's profile picture
Thato Lebethe3 years ago

Would like to learn more

Vadim Savin @notJust.dev's profile picture
Vadim Savin @notJust.dev3 years ago

Tutorial coming soon

Mathew Dony's profile picture
Mathew Dony3 years ago

But isn't this just a ChatGPT wrapper? We can use ChatGPT directly to get movie recommendations.

Vadim Savin @notJust.dev's profile picture
Vadim Savin @notJust.dev3 years ago

Nope, we are not even using openai models. Yes, we are using some NLP models, but we “train” it on our dataset and then we can query it. It can answer question, find similar movies, etc.

˗ˏˋ Terry Sutton ˎˊ˗ — e/shadcn/ui's profile picture
˗ˏˋ Terry Sutton ˎˊ˗ — e/shadcn/ui3 years ago

That's looking great! Any rough patches on the Supabase side there?

Vadim Savin @notJust.dev's profile picture
Vadim Savin @notJust.dev3 years ago

Thanks. It was super easy and smooth to get everything setup with supabase. What exactly do you mean by”rough patches”?

Related Videos