Roasting Your CV

Gemini API React Js Vite Typescript Tailwind CSS

Previously I saw a post from Linkedin that made a roasting to criticize an uploaded CV, curious why did it and how it works. I searched how the process works through the internet and found a way to do it and the time to execute.

To do this I need an AI that can read text from PDF and then provide a response from it, but to get text from the uploaded file I found 2 ways.

1. Image -> Text (OCR)

I tried to convert images from text using an NPM library, but during the process it apparently took longer because it had to read the image then convert it to text and also needed to set the language used to read it. Not to mention if I missed any text.

In the end I didn't use this method, but I was curious, maybe I'll try this on another project.

2. PDF -> Text

The process I chose uses a PDF file to retrieve the text in it, this method is faster as long as the uploaded PDF contains text not images converted to PDF. With this all the text can be read and retrieved directly.

Continue

After getting the text from the CV, the next step is to need AI to process this and then provide the appropriate response. I use Gemini 1.5 from Google for this and to send requests and responses through the API provided. For development using React Js and Tailwind CSS.

This is the response from the one given. hahaha, it's quite good. Don't take it seriously, this is just for fun and also to know what needs to be fixed.

To try it, access here👉 roasting-your-cv.vercel.app