Tushar Kapoor

A self-motivated technical leader.
  • M.Sc. in Computer Science Specialization in Big Data from Simon Fraser University
  • Pro Efficient with python, Spark Streaming, Spark SQL, and machine learning libraries such as TensorFlow & PyTorch
  • Aptitude to independently learn new technologies
  • Completed HacktoberFest 2019

TECHNICAL SKILLS

"Curiosity is, in great and generous minds, the first passion and the last."
        - Samuel Johnson


Libraries

Kafka


TensorFlow


Keras


PyTorch


SciKit-Learn


SparkML


JQuery


Technical Areas

Big Data


Blockchain


Machine Learning


Natural Language Processing - NLP


AWS, Azure & Firebase


Computer Vision


Web Development


Frameworks

Node.js


Flask


Databricks


Hadoop


Map Reduce


Laravel



TECHNICAL EXPERIENCES

"Working hard for something we love is called passion."
        - Simon Sinek

  • TALKSHOPLIVE

             Lead Engineer - Data


    April 2023 - Present


    Revamping data infrastructure with AWS Glue and Athena, fostering data-driven innovation.


    Promotin data engineering best practices, setting industry standards, and ensuring a competitive edge.


    Integrating predictive analytics with AWS SageMaker, boosting customer engagement by 15%.


    Developing a report API via AWS API Gateway and Lambda, halving report generation time.


    • AMAZON

               Software Development Engineer - 2


      March 2022 - April 2023


      Leading the data pipeline automation effort to automate Table and Dynamic creation of ETL pipelines with just a simple UI and click of a button


      Leading effort to optimize Data Set Access Requests generated by external customers


      Part of all the code reviews created by fellow team members


      Developing automated data pipeline using Apache Airflow, AWS Glue and Athena


      Building data flows for machine learning in collaboration with the data science team


      Creating data discovery, data dictionary, and data deletion tool as per the policy requirements, tool facilitates these requirements using a UI which helps customers with a faster access and onboarding experience


      • BEST BUY HEADQUARTER

                 Team Lead (Manager)


        May 2019 - March 2022


        Leading a team of high performing data engineers, responsible for leading the creation of a new data ecosystem (Lakehouse, Data Lake and Snowflake Data Warehouse) at Best Buy


        Developing the team roadmap and performing strategy planning


        Created a Roadmap to migrate Best Buy enterprise data to Azure Data Lake and Snowflake from on-premises legacy systems


        Setting up Best Practices for data engineering discipline (Microsoft Azure, Databricks, & DevOps)


        Leading the project for design and implementation of Cloud Data Lake for streaming/batch/real-time data, for machine learning and near real-time analytics workloads for the entire organization


        Leading the interview and selection process for consultants and full-time employees


        Standing up the Data Engineering Practice at Best Buy


        • IBM

                   System Programmer


          January – June 2018


          Writing JCL ETLs for IBM DB2 database.


          Analyzed, designed and created new JCL JOBs for the maintenance


          Handled performance and capacity management


          • SOPRA STERIA

                     Software Developer Intern


            May – July 2016


            Developed plugin for SonarQube to process Scala which helped the team to successfully process Scala in SonarQube using Java


            Developed an Employee management system for the team, which was developed using Primefaces, Hibernate Framework, and Maven


PROJECTS

Genius is in the idea. Impact, however, comes from action.
-Simon Sinek

  •     E-Ranked: Product Search Tool Using Deep Learning

             Natural Language Processing - Simon Fraser University



      A Deep Learning based search tool using Deep Structured Semantic Model to represent queries and product details in a continuous semantic space


      We use a convolutional-pooling methodology over the Embeddings & contextual window to capture the contextual structure from the query


      Finally, a max-pooling operation is applied to retain the most useful features


    Fall 2019

    •     CRYPTOINTEL: CRYPTOCURRENCY ECOSYSTEM

               Big Data Lab 2 - Simon Fraser University



        An interactive web dashboard of cryptocurrency market for live data analysis, predictions & visualizations using Crytocompare streaming data


        Predictive models are trained using the LSTM and RNN for price prediction and using NLP on news articles and tweets as additional variables


        Sentiment analysis of streaming tweets running real-time on the dashboard


      Spring 2019

      •     REAL TIME OBJECT DETECTION FOR VISUALLY IMPAIRED

                 Machine Learning - Simon Fraser University



          The system detects the objects in front of it in real-time & announces the name of the object using feed from the camera


          Used YOLOv3 and trained it using Keras on darknet53 weights for a manually labelled dataset


        Fall 2018

        •     IMDB - DATA SCIENCE

                   Big Data Lab 1– Simon Fraser University



            Performed ETL on the data of IMDB and Rotten Tomatoes obtained using web scrapping and stored that into Cassandra database


            Used Spark MLlib to make predictions of votes and IMDB score


            Performed various analysis based on several data points


          Fall 2018

          •     AUTOHAUS: A MODEL FOR SECURE & EFFICIENT PARKING SYSTEM

                     Final Year Project – Amity University



              Developed an automatic car parking which is intelligent enough to park the car automatically based on their size and improves the efficiency of current parking systems


              Used combination of various tools and technologies including Java, Arduino, Python, SQL and MATLAB, and used RSA Encryption to make it more secure


            Spring 2017

PUBLICATIONS

"Research is creating new knowledge."
        - Neil Armstrong

    Glimpse into PyTorch3D: An open-source 3D deep learning library - Click to Read

    Object Detector Android App Using PyTorch Mobile Neural Network - Click to Read

    Getting started with Polynote: Netflix’s Data Science Notebooks - 2019 - Click to Read

    Glimpse into Spark 3.0 [Early Access] - Click to Read

    Host a dynamic website on Google Firebase for free using Node.js and Cloud Firestore DB - Click to Read

    Demystifying Random Forest - 2019 - Click to Read

    Tushar Chand Kapoor and Ankur Choudhary, 2018, Image Watermarking using LTP and DCT, Confluence – 2018 - Click to Read

SELF-DIRECTED INITIATIVES

“Life is like riding a bicycle. To keep your balance
you must keep moving.” - Albert Einstein

CO-FOUNDER & WEB DEVELOPER

- Innervate – Startup NGO

- Youth-led organization to bring a positive change in the education in India
- Conducted sessions and helped students to start enjoying the learning process and developed and maintained the website innervate.in
- Students Innervated – 320; Workshops Conducted – 10; Schools Catered – 7;

CO-FOUNDER & DESIGNER

- Papas Inc – Startup

- A venture to provide customized merchandise at the doorstepa
- Oversaw client acquisition & graphic designing
- Identified new business and pinpointed process improvement areas;

Contact With Me

"People Who Are Crazy Enough To Think They Can Change The World,
Are The Ones Who Do." - Rob Siltanen

CONTACT INFO

Send me a message from the panel, or contact via email from below.

Address