Shorticle 38 – Serverless BOT Framework in AWS

Implementing a Chat bot or conversational bot is an important feature that is expected in many industrial usecases like Customer support, Helpline agent, and Front office support systems in Healthcare, Banking, Insurance and Travel industries. Cloud based architectures like Azure AI or AWS Machine learning services or GCP ML services has provided many services suitable for building a BOT service combined with other native cloud services to provide a complete architecture. AWS has even provided pre-build CloudFormation templates from Marketplace to swift develop a Serverless Chatbot service. 

Figure: Example for AWS Serverless BOT Framwork Architecture

A typical architecture of this kind of Conversational Bots using AWS Lambda function and Polly service is shown in below diagram. Lambda is serverless function to implement the business logic; S3 hosts the web application (User interface for Chat bot interface), Cognito for user authentication and security, Polly for getting text to speech conversion for conversational logs (typed in chat box) and DynamoDB to store and retrieve the data logs for chat service. 

Though similar Chat bot service can be developed with Azure or GCP, with AWS CloudFormation template “Serverless Bot Framework” it is easy and simple to develop, customize and deploy a Bot service on the fly. We can also extend this architecture to include Amazon Lex, which is specialized service for conversational chat bots to provide additional facilities of trigger additional events in chat service like sending an email with Standard operating procedure (SOP), creating an incident ticket with backend services or trigger an SNS or SES alerts for an API call.

For further read: https://aws.amazon.com/solutions/implementations/serverless-bot-framework/

#magtechbytes #wipro #shorticle