Posts by Todd Nigro
Solo-built SongForgeAI. Background in engineering + songwriting; writes about craft, rubrics, and what separates competent AI output from the kind someone would record.
8 posts
Narrative Voice: the metric AI lyric tools fail hardest
The fifth rubric essay. Voice is what separates "a breakup song" from THIS narrator’s breakup song — and it’s the single metric where most AI output still sounds like a model writing through a costume rather than a person with a specific background.
Read moreAnatomy of a Forge: one song from prompt to final score
A full walkthrough of what happens between the moment you type a prompt and the moment a finished song shows up on the page. Seven internal phases, two scoring runs, one real example traced from "a heartbreak on a Tuesday" to a 78-composite country ballad.
Read moreEconomy of Language: why the gauntlet cuts lines you love
The fourth rubric essay. Every word is a tax on attention. The gauntlet hates verse 2 line 3 not because it’s bad — but because it earns less than it costs.
Read moreHow I cut 800 lines from a 2,800-line React component
Twenty-five extraction passes, six hooks, eight helper modules, one monolith that almost ate me. Field notes from a refactor that actually shipped.
Read moreWhat "specificity" actually means in a lyric
The word every AI evaluator over-uses and under-defines. A working writer's version: specificity is not detail. It is the detail only this narrator, in this song, could have given you.
Read moreWhy the default score is 50, not 75
Most AI evaluators grade like a flattering tutor. Ours starts every song at 50 and makes the lyric earn its way up. The Gravity Rule, explained.
Read morePublishing the Lyric Scoring Standard v1.0
An open rubric for measuring whether an AI lyric feels alive. 12 metrics, three weighted tiers, four anti-inflation rules. CC BY 4.0. Published.
Read moreThe State of AI Lyrics, 2026
A data essay using the Lyric Scoring Standard as the frame. Where AI lyric output actually lands on the distribution, the six cliché clusters that still dominate, and what 90+ really takes.
Read more