What it is
xjevboost is a Python classifier for tabular data that uses Jev without putting the entire labeled dataset into context. It learns to send smaller slices of the table, asking for more only when helpful, and aims for full-context quality with fewer tokens.
How it uses Jev
Jev classifies each row using a view of selected labeled examples and columns. A greedy tree search learns which views help and when to stop. At prediction time, each Jev answer guides the next step, within call and token budgets.
Primitives:choice
Technique worth stealing
Greedy tree search over context views, using Jev answers to route rows and stop early.
Try it
pip install -e .; export TYPESAFE_API_KEY; use XJevBoostClassifier with JevProvider.
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-centricyes0.65
- Shows a System One patternyes0.63
- Handles uncertaintyno0.05
- Measuredno0.06
- Runnableyes0.62
- Worth recommendingunclear0.56
- Model replicano0.07
- Problem scopescore on a 0–2 scale1.30
- About Jevyes0.98
Signals by Jev jev-1.13.0, card written by DeepSeek V4.1 Flash from the README on 23 Sept 2026.