Overview of Jina AI from a newbie's perspective

Introduction

First of all starting with the most important question,

What is Jina?

So, far with the knowledge that I have accumulated I can sum up the question saying

Jina is a open-source neural-search framework which can be used as your backend to make neural-search projects without having in depth knowledge about neural networks or machine learning.

My Journey

One day while I was exploring the trending projects on Github I came across Jina. But after skimming through the doc once, I thought that this project is out of my scope and I won't be able to contribute anything in this. High sounding concepts like neural networks, neural search,one-hot embedding is being used and being a newbie in the Data Science community I knew nothing about it. But still I thought to myself that what harm would it do if I try the examples given in the docs only for a day. And that one thought changed my assumptions about Jina upside down.

I noticed that using Jina is as simple as writing a print statement in Python (inserts sarcasm). But truly using Jina to make a project is not that hard as it seems in the first look. Jina will basically feel like a drag and drop framework with an amazing documentation to get started with your first project. If you don't believe me (but I hope you do) then head over to their docs and try this after learning some of the basics about Jina like What is Document, DocumentArray, Executor and Flow?.

A little overview about the Fuzzy grep project that I linked above

Have you ever used an extension in your browser called code grepper. If you have then great and for the folks who haven't used this, here's a little summary

Code grepper produces coding snippets based on your search results at the very top of your browser which makes your work a lot faster and easier thus reducing the time spent on Stack Overflow.

So, Jina's Fuzzy grep project is inspired by this and it shows how easy it is to create a project like code grepper in just 30 lines of code.

These type of projects could be considered as the basic for neural search. Now, if you want to know

What is neural search?

Then, in layman's term it can be considered that

Neural search is like asking a human to find something and based on his past experiences he gives you the best solution

In the beginning as I said that using Jina will be like a drag and drop framework that is because Jina is pre-trained with numerous datasets and algorithms which makes it dynamic to plug your own dataset and create an unique project out of it.

So, far after reading Jina's documentation for a couple of days, I have found two bugs and have raised issues regarding that and now, I am working on those to solve the bugs and also I am trying to create a beautiful project with Jina.

For better understanding and to see that how easy it is to contribute to open source, I am providing the links and images of the issues that I raised just by skimming over the docs.

The first issue is here

The second issue is here

This is just the beginning of my journey with Jina and I hope you guys will join me in this journey as well so, that we can make Jina reach its peak.

Goodbye for now, but I will keep updating you guys weekly about my progress with Jina so, stay tuned for that.