AWS Builder Center
Weekend Showcase Challenge: AskAnExpert AI

Weekend Showcase Challenge: AskAnExpert AI

A switchboard for AWS questions, which includes selecting a category and describing what is stumping you. It instantly routes your question to a matching expert, without browsing profiles or guessing who to ask. Confirmation emails are sent to both inboxes before you even close the tab.

Cloud Enthusiast
Vision and What the App Does The AWS landscape takes time to navigate. Getting along with it could be likened to a type of maze. One could ask a large language model for help or even cross-check documentation. Meanwhile, there are lots of experts on community platforms who might be available to help. Besides, there is no quick path from having a networking question to being routed to a networking individual who might be of help. Here is where a most upvoted comment from among the summer weekend challenges, AWS Expert on Demand by Sonal Sarda, comes into play. AskAnExpert AI is a small switchboard, inspired by Sonal's proposal for an on-demand human experts feature. It is a scoped-down version of that concept. You ask a question, pick the category, and state the urgency. It directs the question straight to an experienced builder who has expertise in that specified area. You do not need to guess through user profiles to determine who can answer the question. Both the person asking and the expert receive an email with a ticket ID.
How You Built It A handful of sample experts were fed into a DynamoDB table. Another DynamoDB table was created to store the tickets that have been made so far. The app hosted on AWS Amplify receives the question from the customer, hits the Lambda function, which then scans the DynamoDB experts lookup table, writes a ticket, and fires two emails: one to the customer and the other to the expert. The emails are verified in Simple Email Service.
AWS Services Used Primarily, four AWS services were used. These are listed as follows. AWS Amplify: holds the static frontend. An index.html with the user interface of the entire application was uploaded to GitHub. It was then connected to Amplify and hosted on the cloud. AWS Lambda: contains the function URL. The function holds the routing logic of AskAnExpert AI. Amazon DynamoDB: has two tables. Stores an expert directory with specialty and availability, and a request log of every ticket created. Amazon Simple Email Service: sends the two confirmation emails.
Architecture Overview The flow of the entire app was from the Browser, to the Lambda function URL, to the DynamoDB experts table scan, to the DynamoDB requests table put item, to the two Simple Email Service Emails, to the JSON response rendered as connected in the user interface hosted by Amplify. The browser displays the app with the category, question, and email slots. The Lambda function contains code that triggers the app. The DynamoDB scan finds an available expert in that category. The DynamoDB put item keeps logs of the ticket. The two emails receive the output.
Architecture Diagram

Screenshots
Lambda function

Tables

Request table n DynamoDB

Table of experts

Amplify

App

Request Table With Requests

Emails

First Ticket to Customer

First Ticket to Expert

Second Ticket to Customer

Second Ticket to Expert

Third Ticket to Customer

Third Ticket to Expert

What You Learned
Across the summer, I navigated the AWS console to build apps for the weekend series. I learnt the diverse ways of uploading data in a DynamoDB table, from using the command line interface, to a Lambda function that triggers data upload in a DynamoDB table, through to manual upload of items in the console. Knowing the flow from AWS Lambda to DynamoDB to Amplify and then to DynamoDB again, without the use of a large language model, minimized the process of building the application. Various tests of the function in the lambda interface in the AWS console.
Link to App or Repo
The source code folder can be found here  KathleenCode/askexperts: ask an expert 
Any opinions in this article are those of the individual author and may not reflect the opinions of AWS.
Enjoyed reading this content? Let the author know!

Your likes, comments, shares, and saves help creators reach more builders.

Loading recommendations

Loading article