AITowards Data Scienceabout 2 hours ago
Context Engineering for RAG: Four Typed Inputs
1 min read
Context engineering, named by Tobi Lütke and Andrej Karpathy in 2025, structures single-document RAG into four bricks: document parsing, question parsing, retrieval, and generation. Each brick emits typed pieces that converge on one LLM call with a fixed system prompt for caching.
Level
Hype check
Tap to vote and see what everyone thinks.
