Intro
KYC, or know you customer, is mandatory in financial services - banking, insurance, and accounting. You must validate picture IDs from your customers.
We'll design a Microsoft Power Automate flow that can validate customer IDs (passports and driver's licenses) towards data from a database (Excel).
Learning Objectives
- Using the AI Builder in Power Automate to extract information from identity documents
- Writing WDL expressions to compare data
Description
- Download the course files and extract the data to the Desktop
- Move the KYC Demo to OneDrive
- Move the IDs to Send to C:
- For each ID, validate the data (First Name, Last Name, and Date of Birth) in the Excel sheet Customers.xlsx
- Update the Validated column with Yes or No
- Move the ID file to the correct directory afterward based on the result of the validation
Hints
- It's nice to get familiar with WDL. Don't stress; it's not magic, and we'll get better and better, as we do it in each of the videos 😊
- Knowing a bit about JSONs is nice when we want to manipulate, e.g., file names in Microsoft Power Automate. First, we get the filename, including the extension, then split it into an array by the '.', and get the first element of the array, which is the filename without the extension. Here's our expression for getting the file name:
first(split(triggerOutputs()?['headers']?['x-ms-file-name'],'.'))- Excel Online gives us dates in a five-digit format, days after the 30th of December 1899. So our strategy is to create an expression that adds the five-digit number to the 30th of December 1899, format it with the .NET Custom date and time format, and then we'll have our date in the correct format:
addDays('1899-12-30',int(outputs('Get_a_row')?['body/Date of birth']),'yyyy-MM-dd')
