Skip to main content
pressentation image

Designing the next generation of chatbots

As an AI Engineer at CVS, revolutionized the chatbot experience using generative AI, focusing on creating the best tool for accessing customer support

Visit website
  • AI Model Selection
  • Fine Tuning AI Model
  • Natural Language Processing
  • Front End Development
CVS Chatbot UI during development

The problem

Until 2023, CVS Pharmacy had been grappling with the challenge of accommodating rising customer demands for personalized assistance and insights on medications and health concerns. The existing customer service channels were struggling with the efficient handling of these queries, resulting in potential delays and customer dissatisfaction. To address this, a proposal was put forth to introduce a state-of-the-art Generative AI-based chatbot. This chatbot would swiftly provide accurate responses, empowering customers to effortlessly access information, track prescriptions, and receive tailored health recommendations through natural conversations.

A dramatic ocean scene with lava forming a new land mass.
Generative AI dark based chatbot mobile app concept

Generative AI Based Chatbot

The solution offers substantial advantages, but successful implementation necessitates meticulous attention to data privacy, regulatory adherence, seamless integration, and continuous enhancement. Therefore, several guardrails were created to keep the bot on topic, provide relevant answers from a trusted knowledge base and prevent hallucination.

Easily customizable knowledge base

The most powerful aspect of the Chatbot is the ability to create custom version for any department. The chatbot will only generate answer from is given knowledge base. Out of the box it can be made configurable with minimal effort from developers.

Configuration options for a component.
Configuration options for text.

Next-generation ChatBot experiences

The flexibility and configurability of the bot allowed for developers to create engaging interactive experiences for specific departments within CVS.

Bringing 3D into chatbot

One really cool feature is the 3D screen plugin. Chatbot can explain several concepts using 3D models according to user requests.


Learners can then be directed to specific parts of the model and shown labels. They’re also able to click and drag to orbit around and freely explore at any time.


Users can develop a better visual understanding by displaying custom markers or indicators on the 3D models, according to the users’ requirements based on genarative AI.

Project outcomes

Ultimately the project was successful after the chatbot was adopted across all divisions of CVS to become a foundation for their next generation of customer support.