MongoDB Vector Search Tutorial

MongoDB recently released their vector db offering. If you're trying to develop an AI based product with noSQL, their Atlas Vector Search is super convenient since you can use the DB to store user data alongside embeddings.

Pros
- Super easy to set up
- If you're already using Atlas to store data, no need to configure an extra db just to store vector embeddings
- Free tier will last for a while giving you time to figure out your customers needs

Cons
- Paid plans expensive
- Very recently released so not a lot of info online

[geooogle]

Keep this up sir!! You're quickly showing me the details for tech I don't have time to investigate. You saved me days...

[suryanshsingh]

Hi I wanted to know, can I create this similar index structure without using the cloud version of mongodb atlas .... And just using the plain desktop version of mongodb ?

[LevDiken]

Great video, I’m still learning and the way you explain everything in your videos resonates with me. Thanks Tosh.

[richbennetts]

🔥Totally missed this video! thanks for sharing Tosh!

[jervx]

Can we install mongodb vector locally? I have the mongodb community version installed locally, does that support this vector embedding?

[himanshuahujaofficial]

Thank you so much, Tosh, for the tutorial. It was perfect.

[GaganD-ehrh]

I am encountering an error "Configuration is not a constructor"

[MrJeyOnCrossfit]

This video has so much value ! thanks a lot !

[victorbotteicher]

Thanks so much for this tutorial!

Are you hosting the scraper/embedder locally?

Do you have an idea of how much data is required to adequately train the LLM?

[ramdaneoualitsen]

Hello Thank you for the video. is it possibe to do that without atlas?

[dikawahyu]

Excellent!, thank you very much, my team want to build disease diagnosis chat this video very helpful

[destocot]

(great video, very easy to follow) I had an idea but im trying to think where I could use AI to expand my knowledge.

So lets say i take a bunch of movies, the title, their description, and then embedding in a mongdb atlast
then similar to what you did ill get a query from the user to return the 5 closest movies "e.g. give me a movie with cats in space"

and then i can just return the 5 movies that come back from the nearest neighbor search, but then im asking myself where did i use AI here, so i guess I need to think of a more involved idea

[seragmahmoud]

thamk you a lot.. but iam totaly lost even i completely understand you.. how to make this runs and what about the chunks i will be very pleased if u can provide the whole code or explain a little bit more about them

[MuhammedTan]

This requires Atlas right? Not free version.