AI PM
← Notes
frameworksdata

Rules first, AI for the residual

On Podsque's completeness pipeline, deterministic rules filled ~70% of the gaps at zero cost and the model only handled the ~30% that needed judgment. Reach for AI on the part that's genuinely ambiguous, not the part you could just write down.


[The note "Rules first, AI for the residual" in full — short, conversational, first person.]

The 70/30 split

[Draft section: the 70/30 split. Written when this piece is published — left as a placeholder here rather than fabricated.]

What belongs in a rule

[Draft section: what belongs in a rule. Written when this piece is published — left as a placeholder here rather than fabricated.]

Where the model earns its cost

[Draft section: where the model earns its cost. Written when this piece is published — left as a placeholder here rather than fabricated.]