Nguyễn Đức Long oNddleo

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.

Phân tích và Tối ưu Query

🔍 Phân tích vấn đề

Các điểm nghẽn chính:

ANY operator trên array columns (campaign_ids, tags)
- Không thể sử dụng B-tree index hiệu quả
- Full scan hoặc bitmap index scan tốn kém

Complete Guide: Configure Private S3 Bucket with CloudFront and Cloudflare

Overview

This guide walks through setting up a private S3 bucket with CloudFront as the content delivery layer and Cloudflare as the proxy/caching layer. This architecture provides security, performance, and cost optimization.

Phase 1: AWS S3 Configuration

+-------------+        +------------+        +--------------+
| Client Apps | <----> | WebSockets | -----> | Chat Service |
+-------------+        +------------+        +--------------+
                                                 |
                                                 v
                                         +----------------+
                                         | Message Service|
                                         +----------------+
                                                 |

Importing net/http/pprof adds HTTP endpoints to serve profile files that can be viewed or charted with a command line tool when the runtime.Set* functions are executed.

1. import pprof into the go program

import _ "net/http/pprof"

If your application is not already running an http or https server, add net/http to the program imports and the following code to the start of the main function:

go http.ListenAndServe(":8550", nil)

Technical summary Ultra Scale Kafka system

Core

Apache Kafka
- Usage: Message broker infrastructure, topic management, partition replication
- Key Components: Kafka Streams API for processing, AdminClient for topic management
Apache Avro

Usage: Schema definition, binary serialization/deserialization

Kafka Event Definition for Order Management in Trading

Here's a comprehensive definition for Kafka events in an order management trading system with hash verification for message producers:

{
  "eventType": "ORDER_CREATED",
  "version": "1.0",
  "timestamp": "2025-03-10T08:45:32.215Z",
 "orderId": "ORD-12345678",

	-- ~/.config/nvim/init.lua
	-- Neovim configuration with snacks.nvim terminal and transparent background

	-- Basic settings
	vim.opt.number = true
	vim.opt.relativenumber = true
	vim.opt.tabstop = 2
	vim.opt.shiftwidth = 2
	vim.opt.expandtab = true
	vim.opt.smartindent = true

	SELECT symbol, close, date
	FROM public.historical_data
	WHERE symbol IN ({placeholders})
	AND date BETWEEN %s::date AND %s::date
	AND close IS NOT NULL
	ORDER BY symbol, date

	watch_stylesheet: true
	watch_config: true
	debug: false # Enable debug mode to see more logs

	# This komorebi setting will be used for tray context menu.
	# If you don't use komorebi, you can remove this section.
	komorebi:
	start_command: "komorebic start --whkd"
	stop_command: "komorebic stop --whkd"
	reload_command: "komorebic reload-configuration"