Skip to main content
Bluvy Tube
Press Esc to exit fullscreen
Volume m
/
i
Theater mode t
f
/
RAG, who's that and why should you care? 🧵 Searching a regular DB is easy, grab some SQL and go. But how do you search docs, chats and the rest of the mess by meaning? That's RAG: Retrieval-Augmented Generation

Keyboard shortcuts

Space K
Play / pause
← →
Back / forward 5 s
J L
Back / forward 10 s
0–9
Jump to 0%–90% of the video
Home End
Start / end
, .
Previous / next frame (while paused)
< >
Slower / faster
↑ ↓
Volume
M
Mute / unmute
C
Captions
F
Fullscreen
I
Picture-in-picture
T
Theater mode
?
Show this help

RAG, who's that and why should you care? 🧵 Searching a regular DB is easy, grab some SQL and go. But how do you search docs, chats and the rest of the mess by meaning? That's RAG: Retrieval-Augmented Generation

Sergei Suvorin

Sergei Suvorin

@sergeicode.bsky.social

Subscribe
0 views Published on 22/11/2026 14:21 0:22
RAG, who's that and why should you care? 🧵 Searching a regular DB is easy, grab some SQL and go. But how do you search docs, chats and the rest of the mess by meaning? That's RAG: Retrieval-Augmented Generation
1 comment
?
Chop the text into chunks, turn each into a vector, a fingerprint of meaning, drop it in a vector DB. Question comes in → grab the closest chunks → hand them to the model → it answers from your data
"How do I get my money back" and "refund procedure" share zero words, RAG still matches them
With regular DBs it's text → SQL: you type how many customers churned last month, and the model writes the query itself
Why not dump everything into the model? Context isn't infinite, it costs money, and data changes daily. With RAG you update the base and the model's instantly in the loop
View original post on Bluesky
Link copied to clipboard

Sign in

Time to take a break

You've been watching videos for 30 minutes straight: give yourself 5 minutes away from the screen. Stand up, stretch, grab a glass of water… Your videos will be right here when you're back.