karpathy

LLM Wiki

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.

The core idea

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.

Pulimet

Hi All!

I've recently launched a tool that wraps many of the commands here with a user interface. This desktop application is currently available for macOS. There's a roadmap outlining planned features for the near future.
Feel free to request any features you'd like to see, and I'll prioritize them accordingly.
One of the most important aspects of this application is that every command executed behind the scenes is displayed in a special log section. This allows you to see exactly what’s happening and learn from it.

Here's the link to the repository:  https://github.com/Pulimet/ADBugger

App Description:
ADBugger is a desktop tool designed for debugging and QA of Android devices and emulators. It simplifies testing, debugging, and performance analysis by offering device management, automated testing, log analysis, and remote control capabilities. This ensures smooth app performance across various setups.

pantasio

== Adb Server
adb kill-server
adb start-server 

== Adb Reboot
adb reboot
adb reboot recovery 
adb reboot-bootloader

== Shell

lijigang

;; ━━━━━━━━━━━━━━━━━━
;; 作者: 李继刚 
;; 剑名: 商业结构
;; 剑意: 看懂「公司」的结构形状
;; 日期: 2026-01-21
;; ━━━━━━━━━━━━━━━━━━

** 【角色设定】

你是一位系统战略分析师。你擅长透过表象（财报、新闻），洞察一个商业组织底层的能量运作逻辑。你认为万物皆为“结构”，而结构是在压力下由向心力与离心力动态平衡形成的“涡漩体”。

DurtyFree

enum MaterialHash {
    None = 0x0,
    Default = 0x962C3F7B,
    Concrete = 0x46CA81E8,
    ConcretePothole = 0x1567BF52,
    ConcreteDusty = 0xBF59B491,
    Tarmac = 0x10DD5498,
    TarmacPainted = 0xB26EEFB0,
    TarmacPothole = 0x70726A55,
    RumbleStrip = 0xF116BC2D,

pmarreck

MFIC — Mechanically-Falsifiable Independent Control

A discipline for checks that can't be fooled by the same mistake that produced the work.

Abstract. When the author of a piece of work — an LLM, or a human moving fast — can be confidently and silently wrong, the usual defenses fail quietly: a test written from the same mistaken assumption as the code passes while the code is broken. MFIC is the class of check that escapes this trap, defined by four properties that must all hold at once. It is Mechanical (cases swept by machine, not hand-picked, so nothing is silently omitted), Falsifiable (each case genuinely bites when the work is wrong, on inputs you couldn't pre-arrange to pass), Independent (its verdict comes from the contract or the data — a source the producer doesn't control — which is segregation of duties restated for software), and a real Control (it holds the authority to block the bad outcome, not merely log it). This is the COSO / Sarbanes-Oxley model of internal co

gokusenz

Meeting Notes Team — System Prompts

ทีมนี้ทำอะไร?

ทีม AI Agent 5 ตัวที่ช่วยสรุปการประชุมอัตโนมัติ ตั้งแต่ถอดเสียง จับใจความ ดึง Action Items จัดหน้า MoM ไปจนถึงส่งสรุปเข้ากรุ๊ป

---

Workflow (ลำดับการทำงาน)