Updated: 22-Dec-24
To use the index, click on a section in the left (sidebar).
Note: if you're accessing this on Github, documentation repositories are (mostly) excluded.
Those can be accessed, instead, via my docs site.
Configuration text for an LLM assistant designed to assist users with documenting abusive messages.
Scripts and other programs generated with various LLMs. Sometimes they need refinement. Frequently they just work!
AI-generated project starters for anyone wishing to carry on the ideas!
Configuration for a large language assistant designed to assist with redacting personally identifiable details in documents.
Great online communities discussing use of LLMs including (but not limited to) ChatGPT.
An index of prompting libraries for GPTs, including ChatGPT. Some of these are on Github and others are hosted externally.
Some loosely organised experiments intended to probe the levels of political censorship inherent in LLMs.
Tools, frameworks, and platforms for custom LLM agent creation, including AI workforces/fleets.
Configuration parameters for custom LLM assistants for various tasks along with links to where they are available online.
Configuration for an LLM assistant that performs analysis on sustainability data.
As prompting for data analysis & vis is quite specific, a small side collection of prompts for this purpose.
ChatGPT-generated Python scripts for various purposes.
Some cost estimates for using LLMs via API versus web UI services (like ChatGPT).
Markdown badges for attributing LLM generated (or co-generated) content (using Shields.io).
A few prompts that I am storing in a repo for the purpose of running controlled experiments comparing and benchmarking different LLMs for defined use-cases.
Experiments in evaluating various prompting strategies and LLM performance generally.
Experiment to evaluate the ability of code-gen LLMs to generate a long continuous (single) output.
Simple backup GUI for my Obsidian vault for storing LLM artifacts (prompts, outputs, agent configs).
First entry documentation repository for notes mostly unedited from LLMs.
Some parameteric data about my LLM output repository as it scales.
Interesting GPT outputs demonstrating specific capabilities.
A few system prompts for LLMs intended to improve the format and utility of outputs. Written primarily for LibreChat but should be reasonably frontend-agnostic.
LLM-generated documents describing potential use-cases for GPTs & other LLMs across both professional and business contexts with sections describing potential/predicted evolutions as underlying technology advances.
Documentation for managing prompts and outputs and working professionally with large language models (LLMs).
A quick scaffolding script for populating a repository with folders for organising prompts and outputs, as commonly required when working with LLMs.
A folder structure (scaffold) for those employing systematic workflows for working with LLMs.
Large language models (LLMs) explaining themselves - and how to make best use of them (prompt engineering). Often insightful, though accuracy not guaranteed!
A model/template for a custom GPT/LLM agent whose focus is on helping to ideate and improve a technology idea (example: "Output Hub").
An LLM-generated alias to explore contextually guided prompting strategies.
"High level" documentation/thinking for a system I'm working on for storing large language model (LLM) derived information in a graph-backed knowledge base system.
Some GUIs and scaffolding scripts intended to optimise Obsidian for use as a knowledge base for storing and organising LLM outputs (as well as prompts and custom agent config).
A GUI component for the Obsidian vault project. This saves prompts and outputs into the vault but doesn't integrate with an LLM API.
A utility for running prompts through the OpenAI API and saving them into an Obsidian Vault structured for LLM output storage.
Simple script for creating a Github index using the OpenAI API.
An admin application I'm developing for managing all aspects of working with LLMs professionally and at scale, with functions for prompt logging, prompt library, and custom LLM agent inventorising. Aspiration: add code to notes gradually as project matures!
Streamlit project for generating prompts from templates.
Lists of resources and tools for prompt engineering and evaluation.
A repository of prompts I've used when working with LLMs as well as some example outputs and notes.
Experimenting to test the effect of puncutation in prompts on inference quality.
A vault of over 2,000 prompts and outputs derived from interactions with various large language models (LLMs).
Scripts and utilities for managing custom GPTs built with ChatGPT.
Utility for saving GPT outputs into an organised markdown repository.
Simple text conversion utility. User inputs prompts and outputs, GUI renders to PDF. GPT-authored code!
A markdown note taking application optimise for saving and storing outputs from LLM & GPT interactions (prompts, outputs, agent configurations).
A small and opinionated repository of my thoughts about prompting LLMs to maximise their utility for specific use-cases.
A public notebook of some analytical work with datasets.
Data apps and datasets deployed to Streamlit Community Cloud, Hugging Face, and elsewhere.
Streamlit data visualisation examining correlation between emissions & profitability.
Another data visualisation series comparing companies' GHG emissions with their EBITDA.
(Non-official) derivative version of IFVI Global Value Factors Database in JSON/GeoJSON to facilitate data analysis projects.
Derivative database of IFVI Global Value Factors for data analysis and visualization use cases.
A repository for housing visualisations and analyses of data related to impact investing approaches.
Data analysis and visualisations performed on data from various sources tracking the establishment and progress of outcomes funds around the world.
Streamlit app containing visualisations of the Global Value Factors Database (GVFD) released by the IFVI in 2024.
Utility that creates a human-readable manifest of the relationships contained in a Postgres database.
Utility for helping users understand the classic data relationships in RDMSs (M2M, O2M, M2O etc).
Experiment: trying to create a database schema using natural language.
Seeds a relational database structure for 'Output Hub' output management system.
Little scripts and hacks for working with my favorite database!
My thoughts on a few random topics (for my documentation site), to get a bit of space to jot down things besides tech notes.
Some archived documentation.
Documentation matters! Some of the templates (in markdown) that I like to use.
Short tech notes (as in Github Gists, but versioned in a repo to sync them to a doc site).
A few personal notes jotted down for working with submodules in Github while trying to avoid going insane.
Documentation for derivative database of IFVI Global Value Factors Database (GVFD).
My little tech discoveries available for anyone who wants to read them!
The first seedlings of ideas as they pop out of my head and onto the page/code editor.
Some very basic templates for saving LLM outputs in markdown-friendly notebooks (like Joplin).
Master index for Github repositories indices.
An index to my various documentation projects, both on Github and elsewhere.
A first-entry notepad to jot down some notes that may be developed into blogs (or not). My philosophy and lens on everything from AI to backups.
A list of projects I am working on.
A few formatted markdown files for testing the real-time markdown rendering support of various knowledge management, note-taking and wiki tools.
General purpose wiki for hosting Gists, how-tos, etc.
A small selection of GHG emissions datapoints from various sources.
Calculation tool for "monetising" companies' emissions at various proposed social cost of carbon values.
Data pipeline repository supplying various data visualisation projects.
Calculation simulation utility for monetised GHG emissions.
Calculating monetised GHG emissions for various companies based upon disclosure data.
Experimentary: AI assisted workflows for parsing sustainability reports for target data.
Tool for extracting data-containing pages from sustainability reports.
Source data and documentation for corporat sustainability disclosures/reporting.
Home automation projects that are relevant to people in Israel or Jews generally (e.g. Shabbat and zmanim time integrations).
General Home Assistant dashboards - screenshots and YAML.
Automations and configurations for Home Assistant specifically intended for the observance of Jewish festivals and Shabbat.
A Shabbat/Yom Tov dashboard for Home Assistant utilizing the modifications I've documented elsewhere here.
Little snippets of YAML for HA as I figure out how to make it do stuff. Open sourcing it because .. why not!?
A Home Assistant grocery list optimised for the needs of Israeli consumers.
Data ingress pipeline for Impact Database.
A repository for hosting data about impact investing that is not covered by other more specific repositories here.
A prototype app for displaying impact-weighted accounting data for products serving data from NFC tags (example deployment: a supermarket).
A repository for hosting data and data visualisations related to the work of the International Foundation for Valuing Impacts and other organizations pioneering the development and adoption of these methods (disclaimer: this is an independent project not affiliated or associated with any particular entity).
Experimentary repository; using LLMs to analyse geopolitical developments in Israel & the Middle East.
News monitoring app intended for people living in Israel.
An improved translation of the Home Front Command instructions for finding emergency shelter in Israel.
Jerusalem public shelter locations extracted from ODBL licensed sources. Please do not rely upon these for your own preparedness efforts.
GUIs and other tools that make managing Kubernetes clusters easier for Linux desktop users (I use Ubuntu but will add resources for other distros).
Amazing IDEs that work on Linux distros.
Quick template for creating a local repo with symlinks to key system files on Linux.
A few small learning projects intended at developing very simple GUIs for Fedora Linux.
Quick list of some LLM related apps with Linux desktop support (chat UIs, CLIs, wrappers, etc).
Some scripts for backing up OpenSUSE package lists for expedited reprovisioning.
A list of some extensions for working with large language models (LLMs) within the VS Code IDE.
Markdown documentation for the evolution of my desktop.
A utility for downloading Hebrew fonts from Google Fonts using the API (needs a key).
A few scripts for editing static sites built with the Hugo static site generator.
Public collection of miscellaneous scripts, mostly for automation.
Python scripts that are publicly shared via Github.
Tampermonkey scripts for browser automation tasks.
A model for routing contextual data into vector storage for use with LLMs.
Utility for identifying newly released custom ChatGPTs shared via the ChatGPT App Store.
A Claude-developed, me-prompted incremental BTRFS backup utility with hardcoded serial number verification on the target device.
Ideas and scripts for optimising Github repositories.
Script to generate an index of the Github repos on your local / computer.
Outlining an idea for a personal context repository for enhancing LLM inference.
Notes planning a personal contextual data pipeline.
List of taxonomies generated with GPT for various purposes and open-sourced for ease of reuse.
Markdown badges from the Shields.io project (the ones I like and use frequently).
My public PGP keys.
Periodically updated markdown documentation summarizing manual / non-programmatic methodologies to back up commonly used SaaS / cloud-hosted software utilities.
Various scripts for manually syncing and backing up cloud storage.
Backup utility for Nuclino via the API.
Various Bash scripts, edited configuration packages, and other resources to be manually applied to a fresh Ubuntu installation.
File structure templates for common projects. More specific templates sometimes have their own repository.
Scripts for scaffolding folder structures I commonly use.
Unit exchange calculators for calculations I commonly require.
Streamlit app using Google Cloud APIs to extract core text from a PDF and perform TTS. For listening to long docs. API keys required.
Library of code snippets.
Files that needed to be added to other repositories.
Synthetic data containing personally identifiable information.
For open sourcing the occasional sparks of innovation I have but lack the time to implement them currently.
Scripts for creating taxonomies, mostly Python.
Jewish Shabbat Yomtov and Holidays times and event.
I periodically begin using new tools and services. This is a means of documenting them.
A collection of Hebrew fonts from around the internet.
Loose markdown files documenting various early stage projects (ie, prototypes).
First entry documentation repository for notes mostly unedited from LLMs.
A publicly accessible changelog of all my technical projects.
