Kashif Sohail

Data Science, Machine Learning, Big Data Engineer

Education

MS Data Science

Information Technology University Punjab
2017 - 2019

MicroMasters Data Science

University of California San Diego (edx)
2017 - 2018
Certificate | Transcript

BS Computer Science

COMSATS University Islamabad
2012 - 2016

Languages

  • Urdu (Native)
  • English (Professional)

Interests

  • Swimming
  • Bowling
  • Writing

Career Profile

I am passionate about programming, playing with data is all time my favorite hobby. Making things, processes intelligent and convenient is my ultimate goal. Love to work with real-time systems. Cloud is my life.

Experiences

Data Engineer

Oct 2019 - Now
Telenor Microfinance Bank - Easypaisa
Part of the Data and BI team for Greenfield implementation of DaaS (Big Data Cloud Platform) for Pakistans largest Digital Payment system. Responsible for delivering analytics on the Easypaisa mobile app.

Big Data Engineer

Dec 2018 - Oct 2019
SlashNext Inc., Lahore
Working on the collecting, storing, processing, analyzing and visualization of huge sets of data. The primary focus is on choosing optimal solutions to use for these purposes, then maintaining, implementing, and monitoring them. Responsible for integrating them with the architecture being used across the company.

Research Fellow

Nov 2018 - Aug 2019
Scaled Aviation Industries (Private) Limited
I was working with Scaled Aviation on designing AI Models for Fault Prediction in Light-sport Aircraft Engine. The whole process revolves around gathering, processing and then analyzing millions of records worth data obtained through a large variety of sensors.

Software Engineer

Aug 2016 - Dec 2018
ExpertFlow Pvt. Ltd., Lahore

Product Technical Owner/Lead developer of Generic Connector, the backend behind CRM Connectors, responsible for product design, development, documentation and deployments.

Product Technical Owner/Lead developer of Expertflow Chat Solution, involved in design, code, documentation and deployment support of EF Chat Server & Media Routing Engine. Involved in Chat Bot(s) integration for Hybrid Chat (Customer – Bot – Agent) design.

Part of dev team working on Media Routing Engine for Cisco UCCX, involved in development, documentation, maintenance and deployment support

Intern

June 2016 - July 2016
Microsoft Innovation Center, Lahore

Completed one month "Windows Store Heros" Internship Program offered by Microsoft Innovation Center. Tools and Technologies used: Microsoft Azure, Universal Windows Platform(UWP), ASP.NET MVC.

Microsoft Student Partner

Sep 2015 - Nov 2016
Microsoft, Pakistan

As an MSP my responsibilities include, Run workshops and hackathons Host tech events and give demos on campus Work alongside Microsoft professionals, assisting them during hackathons and events Join exclusive Microsoft training events to learn the latest technologies Grow and nurture a community of students and faculty on-campus and online Enhance their skills and career opportunities

Software Engineer

Sep 2013 - July 2016
Technofeedia

Took a startup to provide business solution to the local business. Lead and develop multiple projects e.g. retail business, schools etc, starting from requirements gathering, architectural design up to development and deployment.

Skills

Programming /Scripting Languages

Java, Python, NodeJS (TypeScript), VB.net, C#, ASP.NET, PHP, Shell/Bash

Databases

SQL Server, MySQL, AzureSQL, MongoDB, DynamoDB, Vertica, GreenPlum

Big Data Tools

Hadoop, MapReduce, ELK Stack, Apache Spark, Google Data Prep, Google Data Studio, Cloud Dataproc, Hive, Pig

Integration

Node-Red, Redis, Apache ActiveMQ, Apache Kafka, Socket.IO, WebRTC

Container Platforms (Deployment)

Dockers

Machine/Deep Learning

Scikit-learn, Google Cloud Datalab, Azure Machine Learning, Jupyter Notebooks, SVMs, Random Forest, Naive Bayes, XGBoost, Regression VGG-16, ResNet, Inception, LSTM, GRU

Cloud Platforms

IBM Bluemix, IBM Watson IoT, Google Cloud Platform, Microsoft Azure, Google Firebase, Cisco dCloud, Heroku, Amazon Web Services

Reporting

Crystal Report, Xtra Reports

Collaboration

Git, GitHub, GitLab, JIRA, Visual Studio Team Services

Projects

Here is the list of Projects which I have completed during my educational and professional career.

Image caption generation using Deep Learning - This project is aimed at building deep learning based merged model to generate Urdu captions for images. We were able to achieve a BLEU score of 0.464 with manually translated Urdu captions on a subset of Flicker8K data. Implemented neural image captioning in TensorFlow, Keras. Changed the architecture of neural network to decrease the training time and improved performance.
Predicting E-Commerce order returns using Machine Learning - Canceled or returned orders can severely impact the profitability of any E-commerce ventures specifically in Pakistan where Cash on Delivery is a common payment method. We have utilized machine learning to predict such orders right in the beginning and avoid costs associated with these orders. Our model can capture 88% of the orders that were eventually canceled. This can give a huge competitive edge to any E-Commerce venture that is struggling with such problems.
Hybrid Chat (Customer-Agent-Bot) Framkework - Expertflow's chat framework allows you to unify chats from customers (FB Messenger, Mobile or WebChat, SMS,..), agents (Slack, Cisco Spark, Microsoft Teams, CRM-specific that tool, Expertflow's multi-chat Finesse Gadget, Cisco ECE Chat,..) and bots (IBM Watson, Microsoft LUIS, Google DialogFlow, rasa.ai,..) handoff a Customer-Bot chat to a contact center agent, a bot can suggest answers to an agent that is chatting with multiple clients ("hybrid chat") See it in action
Media Routing Engine for Cisco CCX - MR Engine routes non-voice tasks (chat, email social media comments and messages) to the most relevant CCX agents while the CCX engine continues to route voice calls as per the Skills, CSQs configured in UCCX; in integration with a variety of market-leading CRMs such as Siebel, SAP, and Microsoft in collaboration with Expertflow/third-party CRM Connectors.
CRM Connector for Cisco Unified Contact Center - A middleware that sits between the telephone system and the CRM system. Using the CTI connectors Agents can place, receive, and transfer customer interactions with full, real-time access to CRM customer data.
Video Chat from your website and Cisco Unified Contact Center - Video Chat is an enhancement of the Expertflow chat framework. It provides a Web-RTC based, Plugin-free, HD quality video channel between contact center agents and your customers. Browsers on both the customer and agent side must support the WebRTC protocol, which includes currently Chrome, Firefox, Opera and Android and iOS. Configure whether the customer should be able to start a video session, or whether this feature should only be available on the agent's request (in case an agent is serving multiple chat sessions in parallel).
Predicting stock price movement Using Machine Learning - Stock Prediction for S&P 500 companies using financial ratios and news sentiments to have a better understanding of investment opportunities.
Enterprise Business Infusion - The Enterprise Business Infusion (EBI) is a cross-platform enterprise resource planning system for the small enterprise based on cloud computing.
Cash on Delivery E-Commerce Portal - Developed and Open sourced responsive Bootstrap E-Commerce Portal in core php. It has fully flexible and dynamic control panel to manage all aspects of website
Smart Information System - Smart information system is a smart school automation system. Its features includes students, faculty records, Fee management, exam and grades management, student's over all behaviour records and reporting this using sms.
ScalingSoft - Weigh Bridge Management Software - A fully automated weight management system for Computerized Weigh Bridges, especially timber market of Pakistan.

Publications

Data Partitioning and Fragments Allocation in Distributed Database

Mar 7-9, 2016
International Conference on Communication System and Network Technologies (CSNT)

Certifications

Docker Essentials: A Developer Introduction

Nov 21, 2018
CO0101EN, provided by IBMDeveloperSkillsNetwork on Cognitive Class

Big Data Analytics using Spark

Aug 23, 2018
DSE230x, provided by UC San Diego on edX

Machine Learning Fundamentals

June 15, 2018
DSE220x, provided by UC San Diego on edX

Probability and Statistics in Data Science using Python

March 21, 2018
DSE210x, provided by UC San Diego on edX

Python for Data Science

Dec 15, 2017
DSE200x, provided by UC San Diego on edX

Honors & Awards

Campus Bronze Medal

Oct 2016
COMSATS Institute of Information Technology

Awarded with Campus Bronze Medal upon successful completion of Graduation with securing 3rd Position in University.

Runner-up in Speed Programing TechnoMoot 2015

May 2015
COMSATS Institute of Information Technology Abbotabad

Secured 2nd Position in Speed Programming competition in a national level TechnoMoot 2015 organised by COMSATS Institute of Information Technology Abbotabad.

Winner in OnSpot Programming, VisioSpark 2014

Oct 2014
COMSATS Institute of Information Technology Wah Cantt

Secured first position in OnSpot Programming Competition in national level event VisioSpark 2014 organized by COMSATS Institute of Information Technology, Wah Cantt.

Training Sessions Delivered

1 Day workshop on Data Science for Everyone

UIIT Arid Agriculture University Rawalpindi

1 Day workshop on Data Science for Everyone

Quid-e-Azam University Islamabad

1 Day hands-on session Practical Data Science using Python

Riphah International University Lahore

1 Day hand-on session Building Realtime Applications using NodeJs

COMSATS University Sahiwal

2 Days hands-on session ASP.NET MVC Quickstart

Superior University Lahore

1 Day hands-on workshop on Introduction to Azure and Cloud

Arid Agriculture University Sahiwal

1 Day hands-on workshop on Web-Based Database Application development

Riphah International University Lahore

4 Days hands-on workshop on C# and SQL Server

COMSATS Institute of Information Technology, Sahiwal

2 Days hands-on workshop on C# and SQL Server

COMSATS Institute of Information Technology, Sahiwal

1 Day hands-on workshop on Microsoft Azure Cloud Platform

COMSATS Institute of Information Technology, Sahiwal

Web Articles

Here is the list of technical articles which I have published on top websites like dzone, csharp-corner and codeproject.

Read files from Google Cloud Storage Bucket using local PySpark and Jupyter Notebooks - This tutorial is a step by step guide for reading files from google cloud storage bucket in locally hosted spark instance using PySpark and Jupyter Notebooks
How to read Compressed CSV files from S3 using local PySpark and Jupyter notebook - This tutorial is a step by step guide for configuring your Spark instance deployed on EC2 instance, virtual machine hosted in cloud or in local environment.
SQL Database Query Editor in Azure Portal: A Deeper Look - Query Editor is a great tool but has very limited features. It has a subset of the functionality of SQL Server Data tools and SQL Server Management Studio.
Dynamic Data Masking In SQL Server 2016 - In this article you learn what data masking is, and how to enable data masking in an existing application and how to give access different users to the masked data in SQL Server 2016.
Populating TreeView up-to N Levels in C# from Database - This article will guide you to “How to populate and add new nodes to TreeView up-to N Levels from Database”. You will be able to add new node at any level and expand this up-to N Levels and generate automatic code for any child.
Fake Band The Microsoft Band Emulator - Fake Band is a library for developing and testing Band apps without connecting to physical Band device.