Thread by @jayjen_x

@pinecone just sold out their OP Stack (Open AI + Pinecone) meetup.

It took them 4 hours to reach 4x the venue capacity.

If you're even remotely interested in AI & GPT, you need to know about Vector Embeddings.

Here's why hundreds of people are rushing to get in 👇

1/6 🧵

A computer doesn't understand words the same way a human does.

LLMs like GPT-4 have their own way of making sense of words (which are technically called 'tokens').

When you say "This tastes like bubble gum!"

GPT will convert it to a list of numbers: [15, 721, 92, ...]

2/6 🧵

This list of numbers is called a "vector embedding".

You can almost think of it like how a computer "understands" your words.

Think of it as: "English in -> Vector Embedding out".

You can try it for yourself with OpenAI's "Embeddings" API.

3/6 🧵

These embeddings are very high-dimensional. Or in other words, they contain A LOT of numbers.

@pinecone is a "Vector Database" for these embeddings.

Think about it like a phone book. Or Dropbox.

But instead of storing phone #s or photos, it stores embeddings.

4/6 🧵

And not only does it store, but it lets you analyze.

Ex) Maybe you have 100 pages of docs for your SaaS.

Each page can be turned into an embedding.

When a user asks a question, you can search "which doc is the most similar to the question they asked"?

5/6 🧵

And Search is only one of *many* possible applications.
There's clustering, anomaly detection, recommendations, and more.

This explanation was extremely simplified, but I hope it gives you a better understanding.

TLDR: If you're building apps with AI, look in to embeddings!

Thanks for reading! 🌸

If you enjoyed this thread:

1. Follow me @jayjen_x for more technical & solopreneur topics
2. RT the tweet below to share this thread with your audience

https://twitter.com/jayjen_x/status/1648324805433741312