jevbooks

← All projects

docjev

jerryjliu/docjev

Document classification and splitting with Jev, LiteParse, and optional LlamaParse.

Classification & Routing67%Runner-up: Data Labeling & Curation

What it is

A Python library, CLI, and local app that classifies a PDF, DOCX, or PPTX against natural-language category rules, or splits a packet into ordered sub-documents with page ranges. Built for document-pipeline developers.

How it uses Jev

Jev predicts the document category or the boundaries between component documents from LiteParse page text. Classification returns one category, probabilities, and review flags; splitting returns ordered categories and page ranges, optionally exporting per-document PDFs.

Primitives:choice

Technique worth stealing

Separate OCR and decision timing, and keep continuation pages together while treating different references as separate documents.

Try it

uv sync && uv run docjev doctor --smoke, then classify or split with rules YAML.

View on GitHub

judged by Jevjev-1.13.0

Evidence

Each line is one question put to Jev about the README. ≥ 0.60 reads as yes, ≤ 0.40 as no; in between Jev is not making a call.

  • Jev-centricyes0.80
  • Shows a System One patternno0.22
  • Handles uncertaintyno0.04
  • Measuredyes0.96
  • Runnableyes0.91
  • Worth recommendingyes0.63
  • Model replicaunclear0.49
  • Problem scopescore on a 0–2 scale1.59
  • About Jevyes0.97

Signals by Jev jev-1.13.0, card written by DeepSeek V4.1 Flash from the README on 21 Sept 2026.