| <?xml version="1.1" encoding="UTF-8"?> | |
| <!DOCTYPE keyboard SYSTEM "file://localhost/System/Library/DTDs/KeyboardLayout.dtd"> | |
| <!-- | |
| Data generated Wed Mar 23 01:35:41 2022 | |
| Generated by kluchrtoxml_64 build 203 | |
| --> | |
| <!--Last edited by Ukelele version 351 on 2022-03-23 at 01:45 (GMT+8)--> |
다음 레퍼런스를 바탕으로, 내 코딩 에이전트 설정을 토큰 효율 관점에서 점검해줘.
목표는 성능 저하를 크게 만들지 않으면서 토큰 낭비를 줄이는 거야.
A pattern for building personal knowledge bases using LLMs.
This is an idea file, it is designed to be copy pasted to your own LLM Agent (e.g. OpenAI Codex, Claude Code, OpenCode / Pi, or etc.). Its goal is to communicate the high level idea, but your agent will build out the specifics in collaboration with you.
Most people's experience with LLMs and documents looks like RAG: you upload a collection of files, the LLM retrieves relevant chunks at query time, and generates an answer. This works, but the LLM is rediscovering knowledge from scratch on every question. There's no accumulation. Ask a subtle question that requires synthesizing five documents, and the LLM has to find and piece together the relevant fragments every time. Nothing is built up. NotebookLM, ChatGPT file uploads, and most RAG systems work this way.
