Skip to content

Instantly share code, notes, and snippets.

View vkuznet's full-sized avatar

Valentin Kuznetsov vkuznet

  • Cornell University
  • Ithaca, NY
View GitHub Profile
@vkuznet
vkuznet / das_client.py
Created October 31, 2012 19:28
New das_client
#!/usr/bin/env python
#pylint: disable-msg=C0301,C0103,R0914,R0903
"""
DAS command line tool
"""
__author__ = "Valentin Kuznetsov"
import sys
if sys.version_info < (2, 6):
# Copyright 2012 Erlware, LLC. All Rights Reserved.
#
# This file is provided to you under the Apache License,
# Version 2.0 (the "License"); you may not use this file
# except in compliance with the License. You may obtain
# a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing,
@vkuznet
vkuznet / ms_config.py
Created June 20, 2017 21:19
Example of micro-service configuration file
"""
MicroService configuration file.
"""
import os
import socket
from WMCore.Configuration import Configuration
# globals
HOST = socket.gethostname().lower()
@vkuznet
vkuznet / mutt.txt
Created December 23, 2020 15:33 — forked from jrziviani/mutt.txt
mutt cheat sheet
Mutt
http://www.mutt.org/doc/manual/manual.html
Select (tag) messages: shift+t (T)
= (string)
~ (expression)
~b expr (message with expr in body)
[b]body
@vkuznet
vkuznet / Makefile
Created February 14, 2021 19:04 — forked from thomaspoignant/Makefile
My ultimate Makefile for Golang Projects
GOCMD=go
GOTEST=$(GOCMD) test
GOVET=$(GOCMD) vet
BINARY_NAME=example
VERSION?=0.0.0
SERVICE_PORT?=3000
DOCKER_REGISTRY?= #if set it should finished by /
EXPORT_RESULT?=false # for CI please set EXPORT_RESULT to true
GREEN := $(shell tput -Txterm setaf 2)
@vkuznet
vkuznet / cookiemonster.go
Created February 5, 2022 23:49 — forked from dacort/cookiemonster.go
Simple script to extract (encrypted) cookies out of Chrome OS X cookie store. Usage: ./cookiemonster domain.com
package main
import (
"code.google.com/p/go.crypto/pbkdf2"
"crypto/aes"
"crypto/cipher"
"crypto/sha1"
"database/sql"
"fmt"
"log"
@vkuznet
vkuznet / httpdumper.go
Created February 7, 2023 17:23 — forked from pramos/httpdumper.go
Go Script to Parse Network or PCAPs and Extract Files from HTTP Streams
package main
import (
"bufio"
"bytes"
"compress/gzip"
"flag"
"fmt"
"io"
"io/ioutil"
@vkuznet
vkuznet / llm-wiki.md
Created May 5, 2026 18:56 — forked from karpathy/llm-wiki.md
llm-wiki

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.