Resume Screener - Extract data from CV PDF documents using UiPath and ChatGPT

[@julietnjura1732]

Wow,it worked!Thanks Reda. Your videos are awesome!

[@redamarzouk]

Glad to hear that :D !

[@dilip.j4152]

what changes must be made in this process if the Ask gpt package is not there ?

[@ibankang]

Do you get any solution of this problem!?

[@880juanjo]

This use is great, I appreciate your videos, I would like to ask, does the use of the GPT chat activity consume nothing at cost, use limit or is it going to expire? I ask this because I have clients who I would like to offer solutions like this, but I would not want to hear later that they tell me that the component stopped working.

[@redamarzouk]

Very good question @880juanjo, According to UiPath documentation the "Ask GPT" activity is not supported so I won't advise you to deploy it for customers as of now. But the upside is that it's free unlike other options, and you can deploy it but inform your client that it's not guarantee to work forever. To have a fully supported extraction project I would suggest document understanding. Finally I tried to answer this in the video here: 23:42

[@mohibkhann]

Good job! We need more educational videos like this :)

[@redamarzouk]

My pleasure! more to come :)

[@MikeOConroy]

Rada great channel well done. Do you have GUI demo model real time in Document Understanding with GenAI extracting queries for say in a back eviro, 1/ best performing branches for the day, 2/ customers whom deposited more than 10K within NY brances 3/ Instructions in real time on how to assist a customer direct via LLM to improve cash flow, ie discount invoicing - Thanks

[@redamarzouk]

Hello Mike, Thank you. I want to understand your use case more to be able to respond. Do you have a document with the 3 information that you mentioned and you want to extract it from them using UiPath? You want to be able to see the extraction of these information in a GUI and control it? If that's the question then you can use Action Center, I have explained everything in this video: kzbin.info/www/bejne/oZiUoZujnKhlgdEsi=hl6BSZ3PJpUn5yy_ If you're asking a different question you can answer under this comment.

[@marcelobarros9807]

What about a pdf with multiple pages with invoices (almost all of them are the same layout), what do I need to use?

[@redamarzouk]

Do you mean all the PDFs are part of the same document or you have multiple documents (multiple CVs) in the same document across multiple pages ?

[@JamshedpurMusafir]

One more question how do I add multiple filters to get files for example i want filenames with extension .docx and .doc I tried Directory.GetFiles(attachmentPath,"*.docx)-> this works for only .docx files but when I do this Directory.GetFiles(attachmentPath,"*.(.docx|.doc)) dosen't work.

[@redamarzouk]

The easiest way to do it is to use the “for each file in folder” activity and inside of it use an If activity with the following statement “currentfile.tostring.contains(“.docx”) or currentfile.tostring.contains(“.doc”)”

[@JamshedpurMusafir]

@@redamarzouk lemme try this. Thank you

[@madhushreekusugal8510]

The activity Digitalise document is not present

[@R0hit024]

i am getting error in the add data row activity. object reference not set to an instance of an object. Please provide me solution

[@redamarzouk]

The data table where you need to add the row should be initialized, watch this video: kzbin.info/www/bejne/Znu2ZZ9tfJ10ebMsi=Lufq_hkSGCzm0e4B

[@farhanraisi9]

Hi, I am not able to find the package "InternalLabs.Gpt.Activites". Can you please guide on this.

[@redamarzouk]

The activity has been replaced by Extract Document Data, Watch this video: kzbin.info/www/bejne/m2rGn2ivjJWEfLs

[@farhanraisi9]

@@redamarzouk thank you so much

[@JamshedpurMusafir]

Well explained. Tried this works with pdf extension CV'S and not with 'docx' format any solution for that?

[@redamarzouk]

In the case of .docx documents. don't use the digitize document I've used. Instead follow these steps: 1.Go to manage packages and install "UiPath.word.activities" 2. use a word application scope activity with the file path 3. use the read text activity inside of it and store it in a string value 4.use that string value like the one I've used in the video inside the askgpt activity. regards

[@JamshedpurMusafir]

@@redamarzouk Thank you for the solution. I rather chose to convert it into pdf in the activity which you suggested. Works nicely.

[@gru6421]

Add Data Row: Object reference not set to an instance of an object. how do i fix this

[@redamarzouk]

There are two possibilities for this error, it's either coming from the ArrayRow or Datatable. For ArrayRow make sure that the syntax is correct: ResponseList(0)("answer").tostring.Split("|"c) For Datatable make sure that the scope of the datatable is "Main Sequence" and you have used the Build Datatable activity correctly Like in the video. Finaly if you still have the problem, debug the process and try to visualise all variables.

[@redamarzouk]

To Download CV Ressources and the process, Join this discord server and head to "ressources": discord.gg/jUe948xsv4

[@hikumar123]

discord link is expired or not working, can you update the resources link please

[@redamarzouk]

Hello @@hikumar123 Thanks for notifying me, here is the updated link discord.gg/jUe948xsv4