JustineAI™ — Medical Malpractice Edition.
Multi-provider medical chronology across 5,000–50,000 pages per case. Causation analysis grounded in cited standard-of-care deviations. Daubert challenge preparation. Certificate-of-merit drafting. Expert-witness management.
MM inherits the full Eve-Legal Operating System — case management, deadlines, conflict check, trust accounting, time tracking, e-filing, e-signature, client portal, Microsoft 365 and Filevine integration. The OS layer is shared across every edition; only the legal-reasoning calibration changes.
This edition is not yet generally available. The page describes the workload it will serve and lets medical malpractice attorneys register interest. We don’t publish dates — we publish the architecture and reach out when the edition is ready.
Eight stages. One Justine. The full medical malpractice lifecycle.
This is the lifecycle JustineAI™ MM is designed to run end to end — the same supervisor pattern that runs personal injury today, recalibrated for medical malpractice practice. Every stage will be auditable, every output attorney-attested work product.
- Stage 01Merit screen & intake
When the edition opens, Justine will conduct structured intake for the prospective med-mal matter — injury mechanism, treating providers, the suspected deviation, and the discovery-rule clock. She is designed to flag whether the facts plausibly state a departure from accepted practice before a single record is ordered.
- Stage 02Multi-provider record assembly
Justine will assemble the full record set — hospital, surgeon, anesthesia, nursing, pathology, imaging, billing — ordering, deduplicating, and chronologizing across providers. Designed to read the entire 5,000–50,000-page record in one 10M-token context, she surfaces sequencing, contradictions, and gaps together rather than chunk by chunk.
- Stage 03Standard-of-care & causation analysis
Each suspected deviation will be anchored to a cited standard-of-care reference — a medical-society guideline or peer-reviewed paper — and matched to the corresponding act or omission in the record. Justine is designed to separate the breach question from causation, isolating where the injury actually traces to the departure.
- Stage 04Certificate of merit
Justine will pre-populate the jurisdiction’s certificate-of-merit or affidavit-of-merit template from the assembled record, citing the specific records and standards in play. The consulting expert reviews and signs; the attorney attests. Designed to meet the statutory pre-suit threshold, not to substitute for the expert’s own judgment.
- Stage 05Expert retention & reports
Justine is designed to manage the expert program — engagement letters, retention budgets, scheduling, and a record digest tailored to each specialty. She will draft expert-report scaffolds grounded in the cited literature for the retained expert to revise, refine, and adopt as their own opinion.
- Stage 06Discovery & depositions
Justine will draft interrogatories and requests keyed to the documented deviations, then prepare deposition outlines for treating providers and defense experts. Designed to cross-reference each deponent’s prior statements against the record, she builds impeachment sequences from the witness’s own words and writings.
- Stage 07Daubert challenge prep
Justine is designed to assemble a Daubert pre-read on each opposing expert — prior testimony, peer-reviewed track record, and methodological gaps measured against the field’s accepted methods. She will surface where an opinion outruns its support, with the supporting authority grounded against CourtListener.
- Stage 08Trial workup / settlement
Justine will model the settlement matrix against venue verdict comparables, life-care-plan costs, and lien exposure, and assemble the trial workup — exhibits, demonstratives, and witness order. Designed to present each figure with its evidence and confidence so the attorney decides the number and the posture.
One Justine. Sub-agents built for medical malpractice attorneys.
The MM edition will run the same architecture as PI: one Justine coordinating stage-specialized sub-agents across the matter, reasoning in a single context. The sub-agents aren’t separately branded — they’re facets of the Digital Employee. These are the ones this edition is designed around.
- Records & chronology sub-agent
Multi-provider record assembly, deduplication, and one-context chronology across hospital, surgical, nursing, anesthesia, pathology, imaging, and billing records — sequencing, contradictions, and treatment gaps surfaced together with the source page behind each.
For example — On a retained-surgical-item matter, the sub-agent will read 31,000 pages across four facilities in one context and reconstruct the operative timeline: the count discrepancy noted in the circulating nurse’s intraoperative record, the radiology read eleven days later identifying the retained sponge, and the gap where no follow-up count was charted — each anchored to its page, ready for the attorney’s review.
- Standard-of-care sub-agent
Breach analysis grounded in cited medical literature — society guidelines, practice parameters, peer-reviewed studies — matched to the specific act or omission in the record. Separates the standard-of-care question from causation; never invents the governing standard.
For example — On a missed-MI matter, the sub-agent will cite the chest-pain evaluation pathway in the applicable cardiology guideline, then point to the discharged patient’s documented troponin order that was never resulted and the ECG never over-read. It will frame the departure against the published standard and leave the breach opinion for the retained cardiologist to adopt and attest.
- Causation sub-agent
Links the documented deviation to the claimed injury — alternative-cause analysis, pre-existing-condition disentanglement, but-for and substantial-factor reasoning, loss-of-chance framing where the venue recognizes it. Distinguishes breach from harm.
For example — On a delayed-sepsis-diagnosis matter, the sub-agent will trace the 14-hour window between the first abnormal lactate and the first antibiotic, model the documented organ-function decline across that window, and isolate the harm attributable to the delay from the patient’s underlying comorbidities — surfacing where the causal chain is strong and where defense will press loss-of-chance.
- Expert & Daubert sub-agent
Expert retention logistics, specialty-matched record digests, draft report scaffolds, and Daubert risk on both sides — opposing experts’ prior testimony, publication record, and methodological gaps, assembled into a challenge pre-read.
For example — On a birth-injury matter, the sub-agent will compile the defense neonatologist’s prior depositions and publications, flag an opinion in this case that contradicts the expert’s own peer-reviewed paper on cooling-protocol timing, and assemble the methodological gaps into a Daubert outline — every cited opinion grounded against a verifiable authority on CourtListener for the attorney to weigh.
The MM workload — described before it ships.
What follows is the architectural commitment for this edition — the workload Eve-Legal F5/reasoner will carry once we open it. The reasoning core is shared with PI. The calibration, taxonomy, and document pipeline are MM-specific.
- Multi-provider chronology across 5,000–50,000 pages
The 10M-token long-context model reads the entire medical record set in one thought — not chunked retrieval. Treatment sequencing, contradictions, and gaps surface together.
- Causation reasoning with cited standard-of-care deviations
Every causation argument is anchored to a published standard-of-care reference (medical society guideline, peer-reviewed paper) and a corresponding deviation in the record.
- Daubert challenge preparation
Opposing experts’ prior testimony, peer-reviewed track record, and methodological gaps assembled into a Daubert pre-read — automatically.
- Certificate-of-merit drafting
State-specific affidavit and certificate-of-merit templates pre-populated from the case record, ready for the consulting expert’s sign-off.
- Expert-witness management
Engagement letters, retention budgets, scheduling, draft expert reports, and Daubert risk assessment — orchestrated by the supervisor.
The same depth of reasoning — on medical malpractice matters.
We proved the reasoning on personal injury. Here is what the same compositional fabric is designed to surface once the MM edition opens — the findings a medical malpractice practice would want caught, each presented with its evidence for the attorney to weigh.
- Record contradiction
Justine will surface — Across a 22,000-page record set, Justine will surface the nursing note documenting a fetal-heart deceleration at 02:14 against the physician’s progress note recording a reassuring strip for the same window — the discrepancy that reframes when intervention was indicated — with both source pages attached for the attorney’s review.
- Standard-of-care anchor
Justine will surface — On a delayed-cancer-diagnosis matter, Justine is designed to cite the published follow-up interval for an indeterminate pulmonary nodule from the applicable radiology guideline, then point to the record showing the recommended six-month CT was never ordered — framing the departure against real literature, never an invented standard, for the retained expert to adopt.
- Causation isolation
Justine will surface — Justine will separate the harm attributable to a medication-dosing error from the patient’s underlying renal disease — tracing the creatinine trend before and after the contraindicated dose and modeling the but-for course had the documented weight-based protocol been followed — surfacing precisely where the causal chain holds and where the defense will contest it.
- Daubert exposure
Justine will surface — Justine is designed to flag where an opposing expert’s methodology departs from the field’s accepted approach — surfacing a prior deposition in which the same expert disclaimed the very technique they rely on here, and a peer-reviewed paper undercutting their error-rate assumption — each grounded against a verifiable authority on CourtListener.
- Damages & life care
Justine will surface — On a catastrophic-anoxic-injury matter, Justine will model future-medical exposure from the life-care plan against venue verdict comparables for similar impairment, layer the Medicare and ERISA lien totals computed before settlement, and present a pre-suit demand and a bottom-line authority figure — each line shown with its evidence for the attorney to set.
- Statute & merit discipline
Justine will surface — Justine is designed to compute the limitations and repose clock under the discovery rule for the venue — surfacing a continuing-treatment tolling argument worth 90 days — and to confirm the certificate-of-merit threshold is met by a named consulting expert before the complaint is filed, with every cited authority grounded against CourtListener.
The AI reasons; the attorney decides.
Tell us about your MM practice.
We onboard new editions in waves. When MM opens, founding-firm slots and beta access go to attorneys who told us about their practice early. The form tags your inquiry as MM automatically — you don’t need to repeat the practice area.
Solo and small MM practices: we will reach out the moment self-serve opens. Mid-size and enterprise: we will schedule a discovery call in the wave-prep window before general availability.