Fixing Typos Without Rewriting the Prompt
Daniel dictates his prompts, catches typos by hand, and keeps running into the same wall: fixing a misspelling is not the same job as understanding what he meant. This episode pulls apart the pipeline into its three real tasks — disfluency removal, mechanical typo correction, and semantic inference — and asks which of them a language model should be allowed to touch. Along the way: why an "um" is a token like any other, why a confidence threshold is a certainty parameter you just reinvented, and why a small task-specific corrector drifts less than a large general model. If you've ever handed…
The skinny
The skinny isn't ready yet — notes appear once the transcript is processed.