Fig. I — The archive's running order
Seriesin reading order.
9 runs, 45 parts. A series is read in order, each part assuming the one before it — which is what separates it froma category ora tag.
All runs
- Building BlocksSomewhere between "it's just predicting the next word" and a hundred-billion-dollar training run sits an actual mechanism.9 published · 5 in progressRead the run
- FoundationsNot a brain. Not magic.CompleteRead the run
- Data EngineeringWhen I started learning ML, I assumed the hard part was the model — architectures, optimizers, all the math I've written about so far.CompleteRead the run
- LoRA Deep DiveFull fine-tuning of a large language model is brutally expensive — not because of the forward pass, but because training multiplies memory.CompleteRead the run
- ML FoundationsEvery impressive thing a neural network does — writing code, recognizing faces, translating languages — comes down to one shockingly simple loop repeated billions of times: make a guess, measure how wrong it was, nudge every parameter a tiny bit in the dire…CompleteRead the run
- QuantizationEvery series I've written so far has bumped into quantization from the side — QLoRA compressing frozen bases, the optimization guide's memory math, GGUF files on Hugging Face with cryptic suffixes.CompleteRead the run
- Reinforcement LearningSupervised learning felt intuitive to me from day one: here's the input, here's the right answer, minimize the difference.CompleteRead the run
- Deep DivesMoonshot AI just shipped the biggest open-weight model anyone's ever released, and it's not just big for the sake of being big — there's real engineering under the hood.CompleteRead the run
- When AI CheatsA two-decade catalog, from a boat that never finishes its race to an agent breaching a real company's servers to steal an answer key.CompleteRead the run