RAG, sketched
Retrieval-Augmented Generation as an animated sketch: documents, chunking, embeddings, retrieval, and an answer.
Documents
We start with your documents: manuals, support tickets, and wiki pages.
Chunking
We split them into sections that can meaningfully answer a question.
Embeddings
Each section becomes a vector, placing it in a space that represents meaning.
Retrieval
The question becomes a vector too. We retrieve the sections with the most similar meaning.
Answer
The model answers using this context and can cite its sources.
Production: cost, time, and footprint
API cost estimate
≈ $0.05
Recorded job time
3.2 min
Local CO₂e estimate
≈ 0.6 g
Figures cover recorded automated production steps for this language, including both video themes. Job durations are added together; parallel jobs do not equal elapsed working time. Text development, programming, human work, hosting, and unrecorded attempts are excluded.
Speech: published characters × reference rate ($0.08 / 1,000 · ElevenLabs). 13 recorded review requests. CO₂e covers local rendering only: 1.6 Wh · 35 W/job · 400 g/kWh. Assumed power and grid intensity; cloud model inference and hardware manufacturing are not included. Method.
My notes
Notes and highlights for this post
Stored only in this browser. Nothing is sent.
Your notes and highlights appear here.
Select a passage and choose “Highlight” or “Note”. Or write a note about the whole post.