The LLM Project: Start Here

Since late 2025 I've been building a language model from scratch in my home office: my own tokenizer, my own architecture, my own data pipeline, trained on my own hardware. Not a fine-tune of somebody else's weights. From-scratch, on a bounded corpus (301 Wikipedia articles about World War II), because a bounded corpus is the only place you can actually measure what a model learned versus what you taught it.

The posts below tell that story in order. If you only read one, read the full story from February. Everything here happened on personal time with personal resources, and everything is real, including the failures, which get named honestly.

The Main Line

November 2025 · 5 min

From N-Grams to Micro-Models: Training AI on a 1080 Ti

Where it started: how far can you get teaching a machine language on an old gaming card? Further than I expected, and not nearly far enough.

January 2026 · 5 min

Does Quantization Make Models "Stupid"? A Home Lab Experiment

Before trusting quantized models as tools in my pipeline, I measured what quantization actually costs. Numbers, not vibes.

January 2026 · 6 min

When Bad Shuffling Breaks Training: A Home Lab Lesson

A sawtooth in my loss curve taught me more about data loading than any tutorial ever did. The bug was mine; the lesson is general.

February 2026 · 12 min

Building an LLM From Scratch

Five months, 35,000 lines of Python, 711 commits: the synthetic data pipeline, the architecture decisions, the failures with honest names, and the model that finally started answering questions.

May 2026 · 3 min

D-Day Quiz: A From-Scratch LLM Passes a Historical Exam

The payoff moment: the model sits a real exam on its domain and passes.

June 2026 · 6 min

Using the LLM Last: Deterministic-First Typography Cleaning

A data-quality philosophy in miniature: exhaust the deterministic tools first, spend inference only where judgment is genuinely needed.

June 2026 · 3 min

My First Dataset on Hugging Face: A 21-Million-Pair WWII Synthetic Corpus

The training data goes public: 21+ million conversational QA pairs across 29 subsets, CC BY-SA, built entirely with permissively licensed models.

June 2026 · 5 min

Automating AI Quality Checks: A Self-Improving Model Loop

The current frontier: a 27B model interrogates my model topic by topic, triangulates every weakness, and writes the problem reports that become the next round of training data. QA that runs while I sleep.

Artifacts

WWII Synthetic Corpus (Hugging Face dataset)

21+ million conversational QA pairs, 29 subsets, CC BY-SA 4.0: huggingface.co/datasets/wayneworkman2012/ww2-synthetic-corpus

The model itself

A from-scratch release is in the works. The current training cycle (probe fleet, failure taxonomy, targeted remediation) feeds it. Subscribe or check back for the release post.

Side Quests

March 2026 · 8 min

EmailLM: How I Solved My Spam Problem with Local AI

The same local-AI toolbox pointed at a household problem.

October 2025 · 9 min

Understanding Prompt Injection: A Hands-On Demo

Adversarial thinking about LLMs, and the mindset that later became the probe fleet.

September 2025 · 6 min

YAML vs JSON: The Hidden Token Tax

Tokenizers bill by their own arithmetic. Measuring the difference paid for itself many times over.

This project is ongoing. The model is being interrogated, measured, and retrained as you read this. New chapters land on the blog as they happen.
Page last updated: July 10, 2026