Data Points in Price Quote AI Model Generation Project

Hello , this is Basil with Agro Accounting .

We're gonna walk through the different steps .

I call it the data points in my price quote A I model generation project .

So there's a couple of items .

The first thing I want to go to is to my website and we're gonna follow along Agro accounting.com .

The website is on a mer N stack .

This is the back end of the website .

This is the document portal .

It is hosted for lack of a better word .

I'm using terms as a layperson on Mongo DB .

So on Mongo DB , I can look up , let's say a client Marcel Paulas , I'm going to copy their name and I'm gonna look them up here and we'll see they had a few work orders too because they worked with me in two separate years .

That's the archive tax year field .

We can get into field details later on for now .

I want you to know that the data is in Mongo DB .

I have a spreadsheet that a developer had prepared for me by extracting the data from Mongo DB and preparing it with Python .

And what we did is a fit of the data from Mongo DB .

I have certain columns hidden here with the data from Drake tax preparation software .

Why ?

There's one column that is critical in Drake .

Our tax preparation program , we provide tax prep services .

That is the start date , the start date will impact the price .

If I go further down , I have a price quote message .

This is data that I entered line by line by line and per our upwork post .

What we're going to do , we're going to look at the price quote amount which is data extracted from Mongo DB .

And we're going to look at the short explanation of the client data .

I want you developer to make a correlation with A I between this dollar amount .

Here's the connection with these words , this dollar amount with these words .

Again , I have quote normalized the data by using as consistently as possible .

The next point to look at on the document portal are the client files , the client , either shares on the left side or through a link .

And this is data that is in Mongo DB .

I believe for the case of Marcel .

Let's look at his I guess 2021 file .

We see documents , link , OK ?

Documents link , admin , documents , link , tax return file , not quite sure how , where that's your specialty , but the documents are here and I want to use an A I supported PDF scanner that can go through the documents , scan them , compare them to the short explanation to the dollar amount and there is a triangulation of data .

There is one more piece to that puzzle and we're gonna go back to the job post .

We said the client's documents share either through , through uh a cloud file or folder link or documents uploaded .

And there's a detailed walk through email for every client I send to the staff .

A detailed explanation .

This is how I want you to prepare the tax return for my client .

This is under an email address .

I have a gmail address and for 2023 work work done this year for the 2022 tax year , we always prepare the year before you can see over here .

For example , for Rachel Resnick , here is an explanation .

She's a new client for our New York office , we file these returns .

These are the forms , this is what you want to look at on the return and I can export this folder .

I have one for 23 1 for 22 work and you would then fit the data where an email , detailed email quote is a piece of data that can be connected .

Let's say with that client's short description , dollar amount and name .

So this is how you're gonna quote , prepare the data .

In a few words , we compare the price do amount with a short explanation with the documents .

We need a robust , strong quality A I supported scanner for that .

And the detailed guidance email and I want you to see the connection between all four points because in the future right now , you're working on historical 21 and 22 prep in the future , you'll do it only from documents .

Thank you guys .