Skip to content

Instantly share code, notes, and snippets.

Transcript Summary Benchmarks

These results are anonymised. The source was one transcript split into overlapping chunks.

Input chunk set

Chunk Characters Tokens
video_transcript00.log 11001 2053
video_transcript01.log 11001 2089
@simbo1905
simbo1905 / probe_ollama_model.py
Last active April 12, 2026 08:50
Probe an Ollama model for generate/chat/thinking support with version safety checks
#!/usr/bin/env -S uv run --script
# /// script
# requires-python = ">=3.11"
# dependencies = [
# "ollama>=0.1.0",
# ]
# ///
"""Probe an Ollama model and log what it supports using the local API.
This script requires `uv` to be on the path.
@simbo1905
simbo1905 / README.md
Created April 9, 2026 04:25
Ollama Tool Calling Benchmark - Lua vs Python vs Bash (MIT License - Simon Massey 2026-04-09)

MIT License - Copyright (c) 2026-04-09 Simon Massey

🚀 We Got Ollama's gemma4:26b Doing Tool Calls! (And Benchmarked the Hell Out of It)

Date: 2026-04-09
Author: Simon Massey
License: MIT


The Mission

@simbo1905
simbo1905 / functional_composition_in_python.md
Last active April 17, 2026 09:39
Typed functional composition in Python for contract-driven pipelines

Emit Functions, Not Frameworks

A truer account of typed functional composition for contract-driven Python

Abstract

The clean way to build Open Data Contracts defined Python pipelines is not to start from a framework, a registry, or a mutable runtime context. It is to start from the actual computation.

@simbo1905
simbo1905 / create_shared_group_macos.sh
Created February 27, 2026 20:53
Generate a new macOS group above a given GID and populate it with specified users
#!/usr/bin/env sh
#
# create_shared_group.sh GROUPNAME MINGID USER [USER ...]
#
# Example:
# ./create_shared_group.sh datameshsolutions 1234 simon consensussolutions
# POSIX sh: no fancy arrays, use positional parameters
if [ "$#" -lt 3 ]; then
printf 'Usage: %s GROUPNAME MINGID USER [USER ...]\n' "$0" 1>&2
@simbo1905
simbo1905 / gux-rfc.md
Last active February 25, 2026 17:07
GUX: GUI UX Description Language

GUX: GUI UX Description Language

RFC Draft 0.2 · February 2026

Status

Draft. Public comment invited.

Abstract

OpenCode Desktop v1.1 on Windows 11 with custom MSYS2 install in user folder

Prerequisites

Download and install MSYS2 into C:\Users\UserName\MSYS2. Confirm the installation has msys2.exe, ucrt64.exe, and other shell variants present. Open msys2.exe directly and run pacman -S git to verify pacman works and git installs correctly. Verify you can use both tools in the shell before proceeding.

Locate your OpenCode v1.1 installation. Find OpenCode.exe at a path like C:\Users\UserName\AppData\Local\OpenCode\OpenCode.exe. Confirm opencode-cli.exe is also present in the same directory. Test the CLI by running opencode-cli.exe web to ensure it boots correctly.

Shut down OpenCode. Close the web interface and any running OpenCode processes.

#!/bin/sh
set -eu
usage() {
cat <<'EOF'
aws_sso_env.sh - Print AWS env exports from an SSO profile
Usage:
bin/aws_sso_env.sh <profile>
@simbo1905
simbo1905 / README.md
Last active January 14, 2026 15:57
Line Histogram - Profile files by line size distribution. Perfect for AI agents, data engineers, and anyone who needs to know if line 847 will eat their entire context window.

📊 Line Histogram — The File Profiler You Didn't Know You Needed

What If You Could See Your Data Before It Floods Your Context Window?

You know that moment when you asked your agent to check the data and before you can tell it not to eat the ocean it enters Compaction.

Yeah. We've all been there.

The Problem

  • 🤯 Huge database dumps with unpredictable line sizes
@simbo1905
simbo1905 / DOCUMENT_STANDARDS.md
Last active January 9, 2026 14:02
RFC: Discoverable and DRY Documentation Architecture for Agents and Humans

RFC: Discoverable and DRY Documentation Architecture for Agents and Humans

Status: Proposed Author: Simon Massey Context: Hybrid Human/Agent Repositories

1. Abstract

This Request for Comments (RFC) defines a strict, normative standard for repository documentation architecture. It addresses the "documentation problem" of staleness, duplication, and high cognitive load for autonomous agents. The architecture mandates a single source of truth, explicit linking (no orphans), and a flat directory structure optimized for discoverability via ls, utilizing naming conventions over directory nesting.

2. Core Philosophy