Productizing AI based call router solution
Customer service via telephone is a difficult task for many reasons. This product uses IBM Watson AI solution component to implement call routing that is part of the most basic and important functions in the call center. With no need for a separate call center team to serve customers in small and medium-sized offices, the AI secretary will start to operate within minutes, and guide customers to the department that best suits their needs. Because this solution product is scalable by using Interactor, it can also be fully utilized in large-scale call center deployment work.
TIME & RESOURCES TO BUILD THE SOLUTION
1 week / 2 engineer
Like IBM Watson AI solution components, it is becoming increasingly easier to use, but still requires a lot of resources to build a custom product integrated with other components such as their CRM and website. No matter how good this AI enabled solution is, it is not easy to decide to invest much budget and time for this new technology.
The AI secretary, which will accept calls on your behalf, will be set up in minutes and call routing will start wherever you want to install. AI enabled call center solution has been a huge project in a traditional way, but not for Interactor. It comes with speed, reliable performance, and scalability.
- AI solution component: IBM Watson Conversation
- Telephony service: Twilio
- Interactor on Cloud: AWS EC2
- http servers: Running on Interactor
- UI front-end for AI based call router
- Other services: Custom REST APIs by Interactor
Development, deployment, and management become all possible in a unified software product. Interactor® was able to build the AI enabled solution with 2 engineers in a week. Interactor brought http server, REST API calls, XML process, interaction logic, and management of AI system all together in declarative ways. This is only the beginning of the AI based call center, and more complicated call center functionalities can be added through Interactor such as chat bot, agent assistant working with client’s data base, and dashboard for call history analysis and monitoring.
HOW INTERACTOR IS USED
|Interfacing:||Interactor® is installed in AWS to interface with various vendor solutions and handle different protocols and data formats.|
|Application Logic:||Solution logic is configured directly on the Interactor®. There’s no additional coding needed, but only Interactor in declarative way. Custom APIs were made for actions of Twilio functions, and services for Watson API calls.|
|Maintenance:||Project had to go through many iterations. Since Interactor® is configuration based, changes are applied with a few commands, bypassing the software development life cycle.|
|Physical connection:||Twilio and IBM Watson are accessed via IP network, and Interactor on Amazon cloud is managing all interface connections.|