A letter from Jim Hughes:

EDU Books has been in the book trade for nearly fiffteen years, selling academic books to some of the largest customers in the world for a decade. Our parent company, Enbee Education Centers, was established in 1997. It is now the largest study abroad and test preparation center in western India with 5 company owned centers in Gujarat. EDU Books operates out of the largest of these centers, a sixty-thousand square foot campus in Vadodara.

In the last two years we have begun expanding from our old wholesale business into two new ventures:

Online retail sales through the Amazon marketplace
Modern institutional distribution in India, Africa and the Middle East.

Our online retail system sells through marketplaces. For now this is just Amazon in the US, but it will be extended to others soon. We collect massive amounts of data on each book we can purchase then use machine learning on collected data to decide what to purchase for inventory. Simple in theory. Hard in practice. We’ve been working on proving the concept and developing the systems to accomplish this for two years. The concept is proven, generating 10 times the standard ROI for marketplace sales on good volume. Our path forward leads to the most advanced purchasing system possible for marketplace sales for nearly any product, including our existing books supply chain. We need good people to build our automated systems which gather, organize and process our data.

Institutional sales happen out of our New Delhi office over email and phone for now. We have an older system for selling direct to institutions online, but don’t use it. We are rebuilding that system in a modern architecture, with all that we have learned regarding metadata, sales data and how the institutions work. The model of institutional book sales online is well established in the West. We are adapting that model for the developing world. This is a long-term process wherein e-books and other digital media can be added to our catalog of available products. There are tens of thousands of schools in the developing world. We are building a system which can serve all of them.

As said before, we need good people. People who can come in, put in real effort, learn the systems and grow with us. We are building a core team which covers several technology stacks. Once established with us you can cross-train among technologies to develop your skills and understanding of our whole business.

This is not an IT service company. We don’t “shoot and ship”. We want dedicated, core people who care about the quality, understand how things work and want to stay around long-term.

Full Stack Programmer

Full stack programmers build our internal controls and our institutional sales channels.

We are using a LEMP technology stack with PHP in a microservice architecture for all our new projects. This is one of the most advanced technology stacks around. The person who set it up for us recently used the same stack in developing the new website for one of the major New York financial firms.

Lumen
A simplified version of the Laravel framework. Laravel is now the most popular PHP framework.
Angularjs
Server side Javascript forms the core of many new sites due to its speed and simplicity.
Nginx
The newest and best open source web server out there. Transitioning from Apache isn’t too hard, and it's worth it.
MySQL
Tried and true, we use this for all our database needs. We are considering using noSQL like Mongo for some things in the future, but all our data is stored in SQL for now.
PHP
We use pure PHP 7+ for small projects, and Lumen is a PHP framework.

Responsibilities:

You would be part of a team responsible for building and maintaining the back-end management and front facing institutional sales websites.

Managing our systems requires:

ERP, order control and warehouse management functions
Controls, interfaces and reporting for the data gathering systems
Interfacing with the Python systems developed by the data engineers
Admin functions for the EDU Books website

All of this is built into a administration site which controls all our back-end operations.

Our front facing institutional e-commerce website is being rebuilt. This is a catalog with millions of products listed. The catalog data all comes from the processed data produced by the Java data systems, and files uploaded into the administration system.

Data Engineer

Our remote team in Bangalore has nearly completed a data collection system in core Java. This system gathers over 750,000 sales and metadata data points per hour, per marketplace, for the 1.5+ million titles we can purchase.

It also processes images and ONIX files delivered by publishers, collects data directly from publisher websites via scraping and transfers the data through the servers to be processed and stored. I should note that we were the first company in India to start using ONIX files. Today only Amazon.in, Flipkart and us use the standard.

We need people to improve and maintain this system. This is a complicated task, spanning several servers and requires strong knowledge of databases, APIs, XML and other technologies.

The Bangalore/New Jersey team consists of a few ten to fifteen year high level architects and programmers from Oracle and other premier technology companies. They will be available for consultation and assistance throughout the work.

Java
We use core Java for almost all our back-end data systems.
MySQL
You'll be dealing with hundreds of gigs of data, so get comfortable with advanced SQL.
Google Drive
All employees get a Google Drive account. We use it for communication, office software and back-up.

Responsibilities:

You will be part of a team responsible for continuing to build and maintain the data gathering and processing systems. This includes ensuring minimal downtime and that the data moves throughout the servers without issue. This is a fully automated system. Once complete the primary task will be ensuring it is always running in proper order.

Data Scientist

We use all the data we gather, Python and machine learning and whatever other tools we can to determine which books to purchase and how to reprice them. As said above, this sounds easy, but it isn’t. Other people are working on this problem too. Long-term, large marketplace selling will be a battle of algorithms and supply. We’re one of the first movers.

We use a multi-model machine learning system, starting with regression and extending through multiple levels. Ideally we’ll be able to wrap the whole system in an optimization process to maximize total ROI. When the EDU Books institutional sales website starts bringing in data we will use that to increase sales through promoting certain books to certain people in that channel.

This kind of position isn’t readily available in India. Certainly not to relatively inexperienced people. A large company will pigeonhole you into a narrow career track, solving the same types of problems in a limited field over and over. We are developing the whole process… figuring out what needs to be done and doing it. You’ll need to learn fast. You’ll need to be dedicated. You’ll need a reasonable history with statistics, algorithms, applied mathematics and business/accounting skills. We’ll hire intelligent training staff with no programming experience, but we’re looking for people who have at least started doing these kinds of problems in Python before.

Python
Easy to learn, hard to master. We spend most of our time working with the pandas, numpy packages.
R
While not used now, we're not opposed to using this for visualization and figuring things out.

Responsibilities:

You will be living in the numbers, figuring out, building and maintaining the brains of our purchasing and repricing systems. We'll be producing reports and programs which are called by the full-stack website and export CSVs. Down the line we'll be feeding data directly into the Amazon API.

Sumit and I want to give your an honest appraisal of the kinds of work we do here, and what you would be doing. We work in an education field, and, given our business needs, must also develop a culture of learning within the company. We all work together on a daily basis. We help each other solve problems. We’re using technology to build a modern company, based around automation and intelligence rather than grunt work. Joining us in this is an excellent move for you and your future.

See you soon,

Jim Hughes
Director of Operations