Skip to content

Instantly share code, notes, and snippets.

View balazsbohonyi's full-sized avatar

Bohonyi Balazs - Zsolt balazsbohonyi

View GitHub Profile

I need to geotag every photo from trip to .gpx track, but GPX Logger accidentaly stopped. Luckly, i has turned on Google location history.

  1. Takeout location history from google.
  2. Extract .json file
  3. trigger the following commands
git clone https://github.com/Scarygami/location-history-json-converter.git

py location-history-json-converter-master/location_history_json_converter.py -f gpx -s 2017-05-06 -e 2017-05-07 "Takeout/Location History/Records.json" "Takeout/Location History/Records.gpx"
@balazsbohonyi
balazsbohonyi / node_create_a_redirect_server.md
Last active July 19, 2025 17:28 — forked from leommoore/node_create_a_redirect_server.md
Create a redirect server in Node.js

Create a redirect server in Node.js

'use strict'

var http = require('http');

var mappings = {
 g: 'http://www.google.com'
@balazsbohonyi
balazsbohonyi / jetbrains_trial_reset.py
Created July 19, 2025 18:10
Jetbeans Trials reset
#!/usr/bin/env python3
"""
JetBrains Trial Reset Tool
A Python implementation that resets trial periods for JetBrains IDEs
"""
import os
import platform
import subprocess
import shutil
@balazsbohonyi
balazsbohonyi / llm-wiki.md
Last active April 11, 2026 20:28 — 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.