New Evaluation — Pharma R&D AI

Dashboard

Submit an LLM output for automated scoring + SME review

1. Select Use Case

PSPV Case Preprocessing

AE Narrative · ICH E2B · MedDRA · SUSAR

Pharmacovigilance case preprocessing — LLM extracts and structures adverse event...

Site Identification

Clinical Ops · Feasibility Scoring

AI-assisted clinical trial site identification and feasibility scoring. LLM summ...

Document Authoring Capability

CSR · SOP · ICH E3 · Labeling

LLM-assisted generation of regulatory documents — CSR sections, SOPs, HAQ respon...

HAQ/RTQ Contract Extension

Health Authority Q&A · HA Queries

LLM-assisted Health Authority Question and Regulatory Technical Query response g...

2. Evaluation Content

Your Name / ID (logged for ALCOA+ traceability)

Query / Input Prompt — what was asked of the LLM

LLM Output * — the text to evaluate

Minimum 20 words required 0 words

Reference / Ground Truth (optional — used for faithfulness scoring)

What happens after submission

Automated scoring runs instantly

Score determines routing

SME reviews borderline cases

Decision logged to audit trail

Score ≥ 90% → Auto-approved | Score ≤ 40% → Auto-rejected | 40–90% → SME review queue

Demo: Load sample output