Will Chen wwwillchen

Anti-hype LLM reading list

Goals: Add links that are reasonable and good explanations of how stuff works. No hype and no vendor content if possible. Practical first-hand accounts of models in prod eagerly sought.

Foundational Concepts

Pre-Transformer Models

any: magic, ill-behaved type that acts like a combination of never (the proper [bottom type]) and unknown (the proper [top type])
- Anything except never is assignable to any, and any is assignable to anything at all.
- Identities: any & AnyTypeExpression = any, any | AnyTypeExpression = any
- Key TypeScript feature that allows for [gradual typing].
unknown: proper, well-behaved [top type]
- Anything at all is assignable to unknown. unknown is only assignable to itself (unknown) and any.
- Identities: unknown & AnyTypeExpression = AnyTypeExpression, unknown | AnyTypeExpression = unknown

Prefer over any whenever possible. Anywhere in well-typed code you're tempted to use any, you probably want unknown.

System Design Cheatsheet

Picking the right architecture = Picking the right battles + Managing trade-offs

Basic Steps

Clarify and agree on the scope of the system

User cases (description of sequences of events that, taken together, lead to a system doing something useful)
- Who is going to use it?
- How are they going to use it?

In this gist I would like to describe an idea for GraphQL subscriptions. It was inspired by conversations about subscriptions in the GraphQL slack channel and different GH issues, like #89 and #411.

Conceptual Model

At the moment GraphQL allows 2 types of queries:

query
mutation

Reference implementation also adds the third type: subscription. It does not have any semantics yet, so here I would like to propose one possible semantics interpretation and the reasoning behind it.

The introduction to Reactive Programming you've been missing

(by @andrestaltz)

This tutorial as a series of videos

If you prefer to watch video tutorials with live-coding, then check out this series I recorded with the same contents as in this article: Egghead.io - Introduction to Reactive Programming.

	<artifacts_info>
	The assistant can create and reference artifacts during conversations. Artifacts are for substantial, self-contained content that users might modify or reuse, displayed in a separate UI window for clarity.

	# Good artifacts are...
	- Substantial content (>15 lines)
	- Content that the user is likely to modify, iterate on, or take ownership of
	- Self-contained, complex content that can be understood on its own, without context from the conversation
	- Content intended for eventual use outside the conversation (e.g., reports, emails, presentations)
	- Content likely to be referenced or reused multiple times

	const webpack = require('webpack');

	require('dotenv').config({
	path: process.env.NODE_ENV === 'production' ? '.env.production' : '.env'
	});

	module.exports = {
	webpack: config => {
	const env = Object.keys(process.env).reduce((acc, curr) => {
	acc[`process.env.${curr}`] = JSON.stringify(process.env[curr]);

	/*
	Infix to postfix conversion in C++
	Input Postfix expression must be in a desired format.
	Operands and operator, both must be single character.
	Only '+' , '-' , '*', '/' and '$' (for exponentiation) operators are expected.
	*/
	#include<iostream>
	#include<stack>
	#include<string>

	# Copyright (c) 2013 Georgios Gousios
	# MIT-licensed

	create database stackoverflow DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;

	use stackoverflow;

	create table badges (
	Id INT NOT NULL PRIMARY KEY,
	UserId INT,

	// The Babylonian Method
	// http://en.wikipedia.org/wiki/Methods_of_computing_square_roots#Babylonian_method
	// @param n - the number to compute the square root of
	// @param g - the best guess so far (can omit from initial call)
	function squirt(n, g) {
	if (!g) {
	// Take an initial guess at the square root
	g = n / 2.0;
	}
	var d = n / g; // Divide our guess into the number