Introducing VERA: A Cited AI Assistant for Digital CMC Regulatory Guidance

• Course summary

  This course follows a webinar that introduces the VERA prototype’s end-to-end workflow, from offline document ingestion and indexing through to real-time question answering, including high-performance vector search and multimodal retrieval of both text and figures (e.g., decision trees and flowcharts). VERA’s metadata filters (region, organisation, document type, year range) that help users quickly narrow results to the most relevant regulatory context are demoed.

  Lessons learned from prototype development and early evaluation are discussed, and next steps for evolving VERA as a continuously updated “living” repository aligned with the DCRL digital tool lifecycle are outlined. Finally, answers to questions from webinar attendees (anonymized) are also covered.

• Course outline

  An Overview of CERSI Training Courses and Course-Specific Learning Objectives

  • Course Introduction

  • Introduction to RAG Systems

  • VERA's Library

  • VERA's Architecture

  • VERA demo

  • VERA Analysis & Next Steps
    Questions & Answers Webinar Clip

  • Ask VERA

• Learning outcomes

  • What VERA is and how it supports understanding of the Digital CMC regulatory landscape

  • How Retrieval-Augmented Generation (RAG) works

  • What is inside the VERA corpus, how these documents are curated, and what queries are out of scope

  • How to use VERA and how filters and multimodal retrieval can help you quickly find the most relevant passages and figures for your queries

• Who will benefit

  Pharmaceutical professionals, regulatory affairs professionals, regulators and researchers who have access to VERA, the Digital CMC CERSI's AI librarian.

This course is delivered by CMAC’s Digital CMC CERSI team.

Available now

Format: E-learning

Free

Approx 1-1.5 hours

Register your interest: