El Ezaby Accelerates Pharmacy Consultations Using Generative AI with IT Visionary on AWS

El Ezaby is dedicated to improving healthcare standards in Egypt and ensuring that its customers receive the best possible support. Across its more than 400 branches, El Ezaby’s in-store pharmacists rely on the expertise of the company’s academy for advice on complex cases and prescriptions.

In addition to their deep knowledge, the academy’s 9 senior doctors—3 of whom are dedicated to providing responses to consultations—have access to a large number of resources through its online learning management system. This resource includes handbooks, previously answered questions, and pharmaceutical databases in English and Arabic. In total, the knowledge base has over 500,000 files of information.

Navigating this huge resource of information to provide responses to consultations could be time consuming. Each query, which could come from any of the company’s hundreds of stores, could take around 2 hours to answer and sometimes up to 24 hours. As the company grew, so did demand for consultations.

“The pharmaceutical field is huge and always evolving,” says Dr. Mohamed Ashry, Elezaby Training Academy. “A high turnover rate among pharmacists further complicates the ability to maintain consistent active expertise across all locations. We needed a smarter way to bring trusted knowledge to our pharmacists right when they need it."

El Ezaby had worked with Cairo-based IT Visionary on previous projects, and the pharmacy company had also hosted its systems on AWS for some years. El Ezaby and IT Visionary decided to investigate how generative AI might be able to provide a solution to speed up the pharmacy chain’s information retrieval process and improve the service that it delivered to its customers. Although the opportunities were clear, there was also concern that generative AI could produce inaccurate results—a particular concern for a company with a focus on good patient outcomes.

AWS developed a proof-of-concept chatbot using Amazon Bedrock, which enables the building and scaling of generative AI applications with foundational models. To improve the accuracy of responses and allay concerns about the AI creating inaccurate responses, AWS and IT Visionary used Retrieval Augmented Generation (RAG) to help the EzBot chatbot comprehend the El Ezaby learning management system. This would enable EzBot to generate precise and context-aware responses based on existing and verified sources of information and data.

The data was built and prepared on AWS infrastructure. With many pharmacists working at each store, the demand on the chatbot could be high. The solution uses Amazon Elastic Container Service (Amazon ECS), to run highly secure, reliable, and scalable containers, and Application Load Balancer to route traffic and enable smooth application operation. AWS Glue is used to manage the extract, transform, and load workloads and enable the chatbot to process large and diverse datasets to deliver near real-time responses.

Testing was extensive, with El Ezaby pharmacists involved at every stage to ensure the AI tool aligned with medical guidelines and real-life pharmacy workflows. The initial results were positive. “Our initial experience with the AI demo provided by AWS was highly encouraging,” says Dr. Ashry. “We saw the potential. We were impressed by its ability to rapidly process and deliver accurate information. However, we also recognize the need to fine-tune the model to ensure the accuracy and reliability of medical information before fully integrating into our operations. We were confident this was a strong step toward smarter, faster support for our pharmacists.”

The testing stage, and the honing of the application, was a collaborative process involving the three companies, with outputs reviewed by the academy’s senior experts over many weeks. But trust in the system soon grew among El Ezaby’s pharmacists, and EzBot was rolled out to the company’s outlets in early 2025.

EzBot is transforming how El Ezaby delivers customer experiences and outcomes in its branches, and the company has reported a 50 percent improvement in user satisfaction scores. Using the RAG-enabled chatbot has resulted in a 70 percent reduction in response times to deliver accurate, context-aware answers. The company’s doctors and pharmacists have embraced EzBot, and the chatbot enables them to deliver faster, more informed guidance. In addition, EzBot fosters continuous learning, keeping pharmacists informed about the latest medical advancements and elevating the standard of care. 

EzBot’s ability to handle complex queries with precise answers grounded in a knowledge base has resulted in a 3-times increase in the number of issues that can be resolved without human intervention. The application can scale on demand to offer seamless operations. Automating routine tasks and queries has resulted in a 25 percent reduction in the customer team’s operational workloads, freeing them to focus more on customers.

The project has been a successful collaboration between the three companies. “What makes me most proud of the project is how effectively it has enabled our pharmacists to deliver faster, more accurate, evidence-based support to our customers,” says Ashraf Moustafa, chief technology officer (CTO) at El Ezaby. “This is just the beginning. EzBot is designed for continuous evolution, and I believe this project will open opportunities for other areas within our organization as well.”