Storage rental rates analyzer.

Product Type: Web Application
Business: Storage Rental
Location: Lehi, Utah, United States.

Company: Stack Storage
Client: Nache Nielson
Website: https://stackstorage.us

The client’s business was about Storage rental services all over the United State. For his business, he needs to keep track of over 10000(ten thousand) locations of the United States from 5 different websites of different companies.

Challenges:

Keeping track of storage rental rates from over 10000(ten thousand) locations of the United States from 5 different websites of different companies on a daily basis is not an easy task and it takes lots of manpower and resources. So the client hires us to build a web application from where he can just log in and see everyday records of storage rental rates from any location all over the US.
And also he would be able to add new location links if he wants. For this project, we need to build 5 different scrapers for 5 different websites which will continuously scrape rental prices and details of that storage every day at once. Each of the websites is built with different technologies so for every website new challenges come up very often. And there was a lot of work with data. Every day new 10000+ data is coming up from the scrapers and the backend system has to manipulate that and sort that in a manner where the user can browse it easily. It was a challenging part to do. But we faced the main challenge in the deployment phase because there were 7 different scripts working together. Maintaining all of them was a challenging and also interesting part to do.

Last outcome:

After dealing with all the difficulties we succeed to deliver the perfect tool to our client that he needs from the very first. We build a web application so that the client can access it from anywhere using his credential. And he can add new location links. The system will automatically detect the website from the link. He can view any location and view the price record of particular size storage from that location. We built a graph system for the data so that the client had a clear view of price changes. And also he can change the graph settings from daily, monthly, and yearly.

How we did this:

When the client came to us with his idea he wasn't sure about the whole thing. We listened to the problem and tried to give him some different solutions. Then we made a proper MVP and started the system design and development process. At first, we build the scrapers for the websites which will go to a link and look for storage and scrap the price and details then save it to a database. Then we build a backend for handling those data for filtering, sorting, and manipulating with a graph model. And also build some APIs for login, adding new locations, editing them, and many more actions. After completing the backend we moved to develop the frontend part. We use graph view for data and list view for locations. There was a clean dashboard and profile page for editing profile details. We deploy the 7 scripts to 2 different servers. For frontend and backend, we use Heroku and for 5 scrapers we use VPS. In the end, we succeeded in solving our client's problem and delivering him the proper tool he needed.

Language & Technologies:

We have used Python as a base. Django for creating the APIs, Selenium for the bot, React and HTML, CSS, Bootstrap for the frontend, MongoDB for the database. We have deployed the Backend and frontend to Heroku. For deploying the scrapers we had to use 2 VPS servers.

Client Review

The Core Devs team was great to work with! Highly likely we will work with them again on additional projects in the future.

Poshmark Automation-SaaS.

Product Type: SaaS
Business: Ecommerce
Location: Gastonia, North Carolina,United States.

Company: Happy Thumbs
Client: Sally Schroder.
Website: www.happy-thumbs.com

The client was a businesswoman who wanted to build a SaaS to provide automation solutions to other business owners for e-commerce called www.poshmark.com.

Challenges:

Poshmark is an e-commerce website where users add their products. And there are lots of functions that a user can use to expand their business. A user has to use those features regularly to keep their business growing. But when the business gets bigger it's tough to do these things manually.
So the client wants us to build an automated SaaS solution for that e-commerce where users can use all the automated solutions instead of doing those things manually.

In this automation industry, we face a lot of challenges while doing a new project. But it is our specialty that we believe our team can overcome any kind of challenge and have the ability to solve unique problems on a daily basis. In this project, a vital challenge was to pass the antibot system of that e-commerce. Now the automation industry is evolving. That's why every popular website focuses on having a strong antibot system. Poshmark was no different. But we have successfully passed that by replicating human behavior while automating the functions. Managing different functionalities at the same moment was a big challenge. For that, we had to build the backend in a manner how it can handle any kind of error that could possibly come after deployment. Managing individual tasks, their times, users, subscription packages was a bit challenging but we managed to do that as well. Also connecting the bot backend with the main backend was a bit tough because of various variables. But we build a strong bot backend that continuously monitors tasks and executes those whose scheduled time matches with time now. And manage to connect the backend with each other.

Last outcome:

After overcoming all those challenges we have successfully delivered the project to the client as a SaaS where she can create subscription packages and regulate users. Users can register, log in, buy subscriptions using PayPal, and schedule tasks for a month. And the bot backend executes the task in the background on a daily basis. After delivering the project the client was so happy with our work.

How we did this:

When the client came to us she just had the idea of this product in his mind. Then we discuss a lot and research on it, how it could be done, what difficulties we may face, and many other aspects of the project. After that, we created a proper specification sheet for the project and got it approved by the client so that we both stay on the same page. Then we started the development process. We had automated the full process of that e-commerce. Build a web application where the admin can log in, create subscription packages, and regulate users. Users can sign up, log in, buy subscriptions using PayPal, schedule tasks at particular times of the day, and pre-schedule tasks for a whole month. After completing the development part to focus on debugging and deploying. At the end of the day, we successfully manage to convert an idea/MVP to a top-class product.

Language & Technologies:

We have used NodeJs as a base. Express for creating the APIs, Puppeteer for the bot, React and HTML, CSS, Bootstrap for the frontend, MongoDB for the database, Paypal SDK, and APIs to build the payment system. We have deployed the Backend and frontend to Heroku. For deploying the bot backend we had to use Microsoft azure.

Client Review

Shuvo has been great to work with and is very responsive and knowledgeable on all things code. Job became larger than originally intended and he was able to handle it smoothly with no issues. I will definitely work with him and his team in the future. Thanks!

Instagram interaction automation.

Product Type: Application controlled using CMD
Business: Social media marketing(SMM)
Location: DüRen, Germany.

Client: Nils Spoelgen

The client was a social media marketer. He prodive various kinds of service to instagram influencers. Such as Story view, Like, follow, etc. He uses a service called JAP for that. The client was from Duren, Germany.

Challenges:

The client and his team need to keep a look on all clients' accounts whether they are posting any new photo or any story or not. If so then they manually go to JAP’s website and place an order based on the content and client type. This was a very lengthy and boring process to follow and it was taking a lot of manpower and resources. So the client wants an automated solution for this.
In this project we had to build a bot which will scan every customer account on instagram on a regular interval given by the client. Automating instagram is not an easy job to do because instagram has a very strong bot detection system. So bypassing their bot detector was a challenge since we have to login first and go to the influencers account one by one to scan if there is anything new or not. If so then the bot had to save that to the database and send a request with the content link to another application which would be placing the order for interaction to JAP using their API service. Using third party API is not anything complex but Instagram automation was the main challenge in this project

Last outcome:

As the client requires, we deliver the tool as a script which can be controlled via CMD.The client can give the inputs such as instagram account credential to use, customer instagram handles, various variables for JAP, etc by CSV files. There were two scripts. One was for scanning instagram accounts, scraping new posts and sending requests to the order placing app, another was for receiving the new content details and placing order for that using the variables was given via CSV files.

How we did this:

Client had a proper vision and a fixed budget for the project. So we need to keep our work in the budget. At first we built the intragam bot which was the main hard job. We build the bot in a manner that it will go to instagram and login to an account given by the client via a CSV file. Then going through all of the influence accounts selected by the client, and scan if they posted a new post or story. If it found any new thing then it will push the content link to a list and after scanning all the accounts it will send a request with the list to the order placing app. After one round it will wait for a fixed time set by the user and then run the scanning process again. In the scanning process if an instagram account got banned which is normal then the bot automatically deleted the blocked account from the account list and used another account to continue the process. We hosted the scripts to a VPS server so that the Client can use it from anywhere.

Language & Technologies:

We have used NodeJS as a base. Express for sending and handling requests, Puppeteer for the bot, and MongoDB for the database. We have hosted two of the scripts on a VPS server.

Client Review

Core Devs has been a great team to work with. During the development process of our tool we ran into a few issues that couldn't be foreseen. Shuvo and his team almost doubled the amount of working hours WITHOUT charging us for that. Great service where you can feel that the developers behind are doing their best to satisfy you.

Our work

Storage rental rates analyzer.

Challenges:

Last outcome:

How we did this:

Language & Technologies:

Client Review

Poshmark Automation-SaaS.

Challenges:

Last outcome:

How we did this:

Language & Technologies:

Client Review

Instagram interaction automation.

Challenges:

Last outcome:

How we did this:

Language & Technologies:

Client Review