The training-data pipeline — adapters, alignment, synthesis, Parquet shards.
View all tagsThe training corpus is the largest single source of leverage in the project. The model can only learn patterns that appear in the data. This article walks through how Mailwoman builds its corpus — what sources are in it, how rows are aligned to BIO labels, and how the synthesis step multiplies the effective size.
The empirical-learner nature of the neural parser
A short article on one DeepSeek API quirk that has burned every mailwoman thread that called it with reasoning enabled. Worth knowing before you write the next one.
This is a roadmap snapshot, current as of May 2026. For current state, see the scope declaration, How Mailwoman parses an address, How Mailwoman resolves a place, and How it works now.
A useful intuition guides Mailwoman's corpus-coverage work: as we add training
Both v0.5.0 corpus threads (B kryptonite and B2 transliteration) used an LLM (DeepSeek) to generate annotated training rows. Both surfaced the same lesson: the substring-match alignment check is structural infrastructure, not a quality filter you can drop later. This article explains why.
The training pipeline turns raw address data sources into a model file that ships on npm. This article walks through each stage end-to-end. The Corpus construction article digs into the first three stages in more detail.