Benoit Jadinon benoitjadinon

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.

Svelte vs React vs Flutter

A int stepper exercise (from this great talk)

The goal is to compare technologies using a reactive form that has a 2-way twist (value can be changed manually, and be changed from different sources) results will be graded in terms of code readabilty and number of redraws needed

svelte repl

react playground

Keybase proof

I hereby claim:

I am benoitjadinon on github.
I am benoitjadinon (https://keybase.io/benoitjadinon) on keybase.
I have a public key ASB-I4oxAigtmDfVCqOjjiXEhPhhQ_ERLILKc6Fu22dfJAo

To claim this, I am signing this object:

	model:
	version: 1
	kind: Table
	columns: []
	pluginVersion: 1.0.0
	summaries:
	Unique: values.unique().length
	Total: values.reduce(number(acc)+number(value), 0).toFixed(2)
	filters:
	and:

	void main() {
	final pairs = [
	'BTC/USDT',
	'BTCD/USDT',
	'BTC/USD',
	'USDT/BTC',
	];

	final search = 'BTCUS';
	final r = search.replaceAll(' ', '').split('').join('.*?');


	//Copyright (C) 2012 Sergey Akopkokhyants. All Rights Reserved.
	//Author: akserg

	/// Emulation of Java Enum class.
	///
	/// Example:
	///
	/// class Meter<int> extends Enum<int> {
	///

	extension StreamExtensions<T> on Stream<T> {

	/// usage:
	/// userStream
	/// //.breakpoint() // to break on ALL events of the stream
	/// .breakpoint(prefix:'user', whenKind(k) => k == EventKind.OnResume)
	/// .listen();
	Stream<T> breakpoint({
	String prefix,
	String message,

	import 'package:cloud_firestore/cloud_firestore.dart';

	import 'jsonable.dart';

	class FireBaseApi {
	final String path;
	final Firestore _db = Firestore.instance;

	CollectionReference _ref;

	class BehaviorStreamBuilder<T> extends StreamBuilder<T>
	{
	BehaviorStreamBuilder({
	Key key,
	BehaviorSubject<T> stream,
	@required AsyncWidgetBuilder<T> builder
	}) : assert(builder != null),
	super
	(
	key: key,

	import 'package:flutter/widgets.dart';

	// (c) Didier Boelens
	// https://www.didierboelens.com/2018/12/reactive-programming---streams---bloc---practical-use-cases/

	Type _typeOf<T>() => T;

	abstract class BlocBase {
	void dispose();
	}