The eWizard ecosystem is powered by four specialized AI agents that drive its intelligence and adaptability. Together, the AI agents automate routine work, reduce the risk of errors, and enable teams to focus on higher-value tasks. By combining advanced language and image models with integrated compliance checks, the AI agents make it possible to create, manage, and reuse content faster and more consistently than ever before.
Each agent focuses on a specific area of the content lifecycle—from answering questions to generating assets and validating compliance—ensuring that every stage of content creation and brand consistency is supported. The main focus of each AI agent is as follows:
With the Answer agent, you can ask questions or start open-ended conversations. By combining a large language model with your integrated knowledge base, the Answer agent provides accurate, context-aware answers that accelerate research and decision-making.
The Content generation agent automatically builds fully brand-aligned interactive emails. It generates both text and image components, allowing teams to create production-ready content faster and with less manual effort.
The Image management agent produces and adapts images based on natural language prompts. Using advanced image-generation and editing capabilities, it helps teams quickly create visuals or update existing assets without leaving the eWizard platform.
The Compliance validation agent performs real-time checks against pharmaceutical marketing regulations and internal quality standards. It ensures that all content remains compliant, reducing review cycles and minimizing compliance risks.
The four AI agents are designed to complement each other as follows:
Streamlining content creation and asset reuse.
Maintaining brand alignment across formats and content item types.
Supporting compliance from the very first draft.
All AI agents connect to the platform default LLM (GPT-4o mini) and to Veeva Vault data sources. They support multiple file formats, including PDF, PPT, and DOCX. For the complete list of supported formats, see the Amazon Bedrock documentation.
Let's go over the main capabilities of each agent in more detail.
Answer agent
The Answer agent is your intelligent assistant for finding, adapting, and refining content with speed and accuracy. Powered by a large language model (LLM) and an integrated customer knowledge base, it lets you ask questions in natural language and receive contextual, compliant, and ready-to-use results. Whether you need direct answers or want to explore ideas in an open-ended conversation, the Answer agent streamlines your workflows and helps you work more efficiently.
In your content management workflow, the Answer agent performs the following core functions:
Information retrieval. Ask questions in natural language instead of relying on complex search queries to get quick, accurate answers.
Content discovery. Find and reuse approved materials from your knowledge base. The Answer agent pulls assets directly from your Digital Asset Management (DAM) system, making it easy to access the right content.
Localization support. Adapt content for different markets while staying compliant with local regulations. Built-in translation capabilities ensure both speed and quality.
The Answer agent supports your daily content work with targeted, AI-driven capabilities with the following key features:
Text and image suggestions from your knowledge base
The Answer Agent searches your knowledge base, populated from the DAM system, to provide relevant, brand-approved text and image assets.
Use case: You request content for a specific brand, and the system returns approved text and images that match your query.
Content localization
Seamlessly translate AI-suggested or manually entered text instantly using the LLM capabilities.
Use case: You select text and request localization, and the system generates a compliant, high-quality translation.
Quality validation
Thanks to the powerful LLM capabilities, refine your text for grammar, tone of voice, and readability.
Use case: You submit a draft for validation, and the system delivers a polished version ready for use.
Smart asset decomposition
Extract relevant content from large collections of documents (HTML, PDF, and other approved formats).
Use case: You request brand-specific text and images, and the system pulls and assembles the right content from unstructured sources in your vector knowledge base.
Reference linking
Get direct references to source documents when retrieving content.
Use case: You generate brand-related text and images, and the system provides approved assets along with references to the source documents they were derived from.
Content generation agent
The Content generation agent helps you create interactive content—covering both the layout and assets—for emails and email briefs. You can provide a simple prompt describing the goal of your email, and the agent generates complete layouts that combine text, images, and modules. All suggested content is sourced from your knowledge base or DAM, enriched with AI where needed, and styled according to your master template to ensure brand consistency.
The Content generation agent streamlines the design process by eliminating repetitive setup tasks and giving you brand-aligned content ideas right from the start.
To simplify content creation and ensure brand consistency, the Content generation agent provides the following core functions:
Asset sourcing. Retrieve assets directly from your DAM, AI, or a combination of both.
Brand styling. Automatically apply master template headers, footers, and styles to all generated content.
Image management agent
The Image management agent helps you create and refine visuals directly through natural language prompts. It's designed to support content creation when approved DAM resources are unavailable or insufficient, while still maintaining brand alignment and compliance.
Behind the scenes, the agent leverages a scalable architecture that can work with different generative AI models for image creation.
The Image management agent is responsible for the following core functions:
Image generation. Generate entirely new images based on natural-language descriptions tailored to your content needs.
Image tuning. Adjust or refine previously generated visuals to better match your requirements.
Image localization. Adapt visuals for different markets by reworking an image while preserving key parameters.
Compliance validation agent
The Compliance validation agent together with the MLR Acceleration agent safeguards your content by performing real-time checks against pharmaceutical marketing regulations, quality standards, and approved references. Its goal is to ensure all materials remain accurate, compliant, and ready for use in highly regulated environments.
The following core functions define the agent value:
Reference validation. Verify that all claims are supported by relevant, approved references.
Grammar quality. Detect grammar, spelling, and language quality issues across several languages, including support for pharmaceutical terminology.
Asset verification. Confirm that all content assets are sourced from approved DAM systems or claims libraries.
Regulatory compliance. Automatically evaluates content against international and local pharmaceutical marketing codes (PhRMA, EFPIA, ABPI, PMCPA, etc.) with country-specific rule mapping.