Use Vectorize to add additional context to your AI Applications through RAG -Gamertown

Craig uses Vectorize to store embeddings of the IGDB (Internet Game Database) from Twitch. Using specific context helps add knowledge to your Workers AI chat applications through Retrieval Augmented Generation, RAG using Llama 3.1.

App: https://gamertown.craigsdemos.workers...
Repo: https://github.com/craigsdennis/gamer...

Create an account on Cloudflare today for free: https://dash.cloudflare.com/sign-up

Tools mentioned:
https://developers.cloudflare.com
Workers AI: https://developers.cloudflare.com/wor...
Vectorize: https://developers.cloudflare.com/vec...
IGDB API: https://api-docs.igdb.com/#getting-st...
Embedding Model: https://developers.cloudflare.com/wor...
98.css: https://jdan.github.io/98.css/
Oddworld Abe's Exodus: https://www.igdb.com/games/oddworld-a...


Chapters:
00:00 Intro to Gamertown
00:32 Large Language Model Gamertown Demo
02:30 RAG explanation
03:07 Vector database
04:04 Demo using Vectorize
04:55 Review the code
05:40 Introducing standalone question concept
06:28 Create an embedding
06:51 Use Vectorize to query existing embeddings
07:40 Create the additional context for the System Message
08:35 Wrap up

#cloudflare #cloudflareworkers #developer #vectorize