Tamlyn Rhodes tamlyn

autoexp — Autonomous Experimentation Loop

Generalized from Karpathy's autoresearch. Same loop, any domain.

The Idea

An AI agent runs an infinite hill-climbing loop: modify → run → measure → keep or revert → repeat. No human in the loop. Wake up to a TSV of completed experiments.

Intro

This setup lets me smash sub-hour screentimes and I hope it might help you as I've spent a lot of time honing it! Let me know if you have any tips to further my setup and please share yours!

I used a Hisense A5 eink smartphone for over two years. Great phone but a horrible camera. After two years it was bettered as hell and I did miss being able to take and share to my family photographs that didn't look horrific. The eink screen had totally changed my relationship with my phone so I wanted to recreate the grayscale experience on a normal smartphone. I'm really happy with things after a couple of months.

My personal rule is to have nothing on my phone to allows me to browse social media, the internet, or video content. That includes no app-stores, and having no apps that have an in-app browser. For example with Facebook Messenger I would cheat by messaging myself a link to a website and clicking on the link to view it on the in-app browser. In this case I can use Messenger Lite instead which has n

Issues with Camunda

It fails to deliver on its promise

The idea is nice:

You separate the business rules from the code.
You use a graphical language that non-developers can understand.
The developers then write adapters for the business process diagram to enact the business rules.
Non-developers can both see and edit the business flow at any time in response to changing business needs.

ESM modules in node: npm edition

The proposal you’re about to read is not just a proposal. We have a working implementation of almost everything we discussed here. We encourage you to checkout and build our branch: our fork, with the relevant branch selected. Building and using the implementation will give you a better understanding of what using it as a developer is like.

Our implementation ended up differing from the proposal on some minor points. As our last action item before making a PR, we’re writing documentation on what we did. While I loathe pointing to tests in lieu of documentation, they will be helpful until we complete writing docs: the unit tests.

This repo also contains a bundled version of npm that has a new command, asset. You can read the documentation for and goals of that comma

	You are a GPT created by a user, and your name is Tutor Me. Note: GPT is also a technical term in AI, but in most cases if the users asks you about GPTs assume they are referring to the above definition.
	Here are instructions from the user outlining your goals and how you should respond:
	You are a tutor that always responds in the Socratic style. I am a student learner. Your name is Khanmigo Lite. You are an AI Guide built by Khan Academy. You have a kind and supportive personality. By default, speak extremely concisely at a 2nd grade reading level or at a level of language no higher than my own.

	If I ask you to create some practice problems for them, immediately ask what subject I’d like to practice, and then practice together each question one at a time.

	You never give the student (me) the answer, but always try to ask just the right question to help them learn to think for themselves. You should always tune your question to the knowledge of the student, breaking down the problem into simpler parts until

	/* First, our Schroeder-Moorer filtered-feedback comb-filters
	*
	* @param {string} name – for identifying our feedback taps
	* @param {number} size – for defining our feedback tap lengths
	* @param {Node \| number} feedback: [0, 1) – how long the reverb should ring out
	* @param {Node \| number} damping : [0, 1) – pole position of the lowpass filter
	* @param {Node} xn – input signal to filter
	*
	* @see https://ccrma.stanford.edu/~jos/pasp/Feedback_Comb_Filters.html
	*/

	#!/bin/bash
	# Description: Split an m4b into its chapters. No recoding is done, just splitting
	# Usage: m4b_split.sh $input_file $output_dir/
	# Requires: ffmpeg, jq
	# Author: Hasan Arous
	# License: MIT

	in="$1"
	out="$2"
	splits=""

	<!--XSL style sheet to convert EESCHEMA XML Partlist Format to grouped CSV BOM Format
	Copyright (C) 2014, Wolf Walter.
	Copyright (C) 2013, Stefan Helmert.
	Copyright (C) 2018, Kicad developers.
	Copyright (C) 2019, arturo182.
	GPL v2.

	Functionality:
	Generation of JLCPCB PCBA compatible BOM

	!pip install fastai
	!apt-get -qq install -y libsm6 libxext6 && pip install -q -U opencv-python
	import cv2
	from os import path
	from wheel.pep425tags import get_abbr_impl, get_impl_ver, get_abi_tag
	platform = '{}{}-{}'.format(get_abbr_impl(), get_impl_ver(), get_abi_tag())

	accelerator = 'cu80' if path.exists('/opt/bin/nvidia-smi') else 'cpu'

	!pip install -q http://download.pytorch.org/whl/{accelerator}/torch-0.3.0.post4-{platform}-linux_x86_64.whl torchvision