Back in time, in a startup project, the request was explicit: “I don’t want the customer care hotline ringing.” We had some regulatory items to accomplish like call recordings, long-term retention of multiple systems, and high availability of all systems.
So, once everything was set on AWS, I had the idea to use Amazon Connect. Creating the environment and integrating it with other AWS services like S3, Polly was innovative.
The idea was to create an Amazon Connect flow with fundamental interactions, validations on our user database, record everything, and process documents. Amazon Connect is already a great product, and its integrations with AWS products and third-party companies bring reliability and agility to the business. But the most significant driver was how to process documents and add intelligence with cutting-edge technology.
Amazon Connect: Connect is an easy-to-use omnichannel cloud contact center that helps you provide superior customer service at a lower cost.
Amazon Transcribe: Amazon Transcribe makes it easy for developers to add speech-to-text capabilities to their applications.
Amazon Comprehend: Amazon Comprehend is a natural-language processing (NLP) service that uses machine learning to uncover information in unstructured data.
Amazon Quicksight: Amazon QuickSight is a scalable, serverless, embeddable, machine learning-powered business intelligence (BI) service built for the cloud.
AWS Lambda: AWS Lambda is a serverless compute service that lets you run code without provisioning or managing servers, creating workload-aware cluster scaling logic, maintaining event integrations, or managing runtimes
AWS S3: Amazon Simple Storage Service (Amazon S3) is an object storage service that offers industry-leading scalability, data availability, security, and performance.
The description of High-Level Design follows:
- Amazon Connect processes call.
- The call ended, the recording is stored in an S3 bucket.
- A lambda trigger invokes Amazon Transcribe API
- After Transcribe process the document, the result is stored in another S3 bucket.
- The stored transcribed document will invoke another Lambda that will call Amazon Comprehend API.
- After Comprehend process, the document will store the results on a Data Store S3 bucket.
- The results will be consumed by Amazon Quicksight dashboards that will give insights to enrich data on your site, reducing calls.
Needless to say that encryption, logging and monitoring must be put in place. From this HLD, the enhancements can run through the infinite. Things like NPS evaluation that could evolve into a gamification platform for the best problem-solvers, create ML models to drive the customer to the right choice without humans, a solid and detailed knowledge base…