How the EL edition is designed to work.
JustineAI™ EL runs the same compositional fabric as the in-market PI edition — Eve-Legal F5/reasoner, one Justine supervising stage-specialized sub-agents in a single 10M-token context. What changes is the calibration. Each deep-dive below describes a mechanism this edition is built to carry once it opens. The architecture is real today; the EL calibration is on the roadmap.
- 01Ingestion
Reading the whole class record in one context
Payroll, timekeeping, and personnel files for thousands of class members, reconciled in a single reasoning pass.
Read deeper → - 02Damages
Modeling class-wide damages under the right multiplier
Fluctuating-workweek versus time-and-a-half, liquidated damages, and statistical sampling, computed across every member at once.
Read deeper → - 03Certification
Engineering Rule 23(b)(3) predominance
Building the common-question record and pressure-testing it against the individualized-inquiry defense before the employer raises it.
Read deeper → - 04Collective
Orchestrating FLSA collective notice and opt-in
From conditional certification to a managed opt-in population — notice, consent tracking, and the limitations clock per member.
Read deeper → - 05Briefing
Drafting the certification brief from the record up
Rule 23 and FLSA similarly-situated arguments assembled from the reconciled record, every factual claim cited to source.
Read deeper →