Kurukshetran Kurukshetran

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.

Performant local mixture-of-experts CPU inference with GPU acceleration in llama.cpp

Introduction

So you want to try one of those fancy huge mixture-of-experts (MoE) models locally? Well, whether you've got a gaming PC or a large multi-GPU workstation, we've got you covered. As long as you've downloaded enough RAM beforehand.

Anatomy of a MoE Model

MoE models are described in terms of their total parameters and active parameters - i.e. DeepSeek V3 671B A37B has 671B total parameters, but we are using only 37B parameters at a time during each forward pass through the model.

how to leverage oracle's temping offers

free tier limits

The limits of the free tier say that you can create up to 4 instances.

x2 x86 instances (2core/1g)
x2 ampere instances (with 4core/24g spread between them)
200GB total boot volume space across all intances (minimum of 50G per instance)

	#!/bin/bash

	#https://becomegeeks.com/blog/how-to-decrypt-a-pdf-file-on-mac-remove-password-from-pdf-files-on-macos/

	PASSWORD=""
	INPUT_DIR=""
	OUTPUT_DIR=""

	while [[ "$#" -gt 0 ]]; do
	case "$1" in

	<!--
	CHECK IF A STRING IS NULL OR EMPTY
	-->

	<!-- Example XML -->
	<group>
	<item>
	<id>item 1</id>
	<CategoryName>blue</CategoryName>
	</item>

	package main

	import (
	"crypto"
	"crypto/rsa"
	"crypto/sha1"
	"crypto/x509"
	"encoding/base64"
	"fmt"
	)

	package main

	import (
	"bytes"
	"crypto/rand"
	"crypto/rsa"
	"crypto/x509"
	"encoding/base64"
	"encoding/pem"
	"errors"

	package main

	import (
	"bytes"
	"crypto/aes"
	"crypto/cipher"
	"encoding/base64"
	"fmt"
	)

	> This software/architecture includes concepts and logic developed for Matryca Brain.
	> Original Author: Marco Porcellato - Italy
	> First published: June 21, 2026.

	package main

	import (
	"net/http"
	"regexp"
	"strconv"
	"time"

	"github.com/davecgh/go-spew/spew"