Close

Parth Nagarkar

Software Engineer

Download Resume

   

About Me

A Software Engineer at Amazon Web Services (AWS) with an MS in Computer Science from Stony Brook University. Specializing in backend development, I'm currently a part of the AWS Supply Chain product team, focusing on the Demand Forecasting module. My hands-on experience ranges from architecting scalable systems to collaborating with diverse teams to ensure efficient data processes. I bring expertise in multiple programming languages and frameworks, including Python, Java, Spring, Flask, and AWS. My stint at Amazon has fortified my knowledge in supply chain systems, and I take pride in influencing architectural decisions and upholding AWS standards.

Experience

Larsen & Toubro Infotech

Software Engineering Intern

  • Worked in an Agile team of .NET developers and designed an Ecommerce website for Handicraft products.
  • Implemented a live visit counter to count the number of daily unique visitors on the website using HttpCookie.
  • Technologies: ASP.NET, HTML5, CSS3, jQuery, Bootstrap, MS SQL Server.

TryCatch Group

Machine Learning Intern

  • Developed a URL classification system using Python for scraping data to classify URLs as benevolent or malicious.
  • Performed preprocessing and feature engineering using Numpy, Pandas and trained machine learning models using Sklearn.
  • Built web crawler to scrape URLs from a website using BeautifulSoup and deployed it on AWS by using Flask in backend.
  • Enhanced feature generation performance by implementing caching and multiprocessing to speedup execution by 30%.
  • Technologies: Python, AWS, Flask, Numpy, Pandas, BeautifulSoup, Sklearn, NLTK.

Hiranandani Group of Companies

Data Science Intern

  • Implemented web scraping module of a project to extract Twitter data to identify and connect with potential real estate buyers.
  • Engineered Python scripts to scrape data from Twitter using Twitter API, BeautifulSoup and clean the data using NLTK.
  • Designed a multi-staged method for data extraction, ranking the prospective buyers and scraping their contact information.
  • Leveraged multithreading and parallel processing using tools like Google Colab to optimize execution time by 40%.
  • Technologies: Python, SQL, Twitter API, Pandas, Numpy, NLTK, Google Colab, BeautifulSoup, REST APIs.

Stony Brook University

Graduate Teaching Assistant

  • Working as a teaching assistant for an undergraduate course named Fundamentals of IT (ISE-218) having 150 students.
  • Executed all teaching assistant duties including mentoring, lecturing, researching and clerical help.

Stony Brook University

Graduate Teaching Assistant

  • Working as a teaching assistant for an undergraduate course named Cloud Computing (ISE-390) having 100 students.
  • Executing all teaching assistant duties including mentoring, lecturing, researching and clerical help.

Stony Brook University

Research Assistant

Amazon Supply Chain

Software Engineering Intern

  • Built a service to compare the buying plans for inventory procurement generated by legacy system and new system to validate the performance of the new system before release to production.
  • Implemented a design to set up Amazon SQS and transfer generated plans from the two systems to comparison API.
  • Engineered an API design to receive the generated plan data, parse it and store the data Amazon DynamoDB table.
  • Incorporated JUnit and Mockito unit testing frameworks to achieve line coverage of 95% and branch coverage of 98%.
  • Developed the comparison logic to compare the plans and send them to ElasticSearch for visualizing it using Kibana.
  • Technologies: Java, Spring, TypeScript, NoSQL, Amazon DynamoDB, Amazon SQS, Amazon OpenSearch, Elastic Stack

Jefferson Health

Software Developer

  • Spearheaded the resolution of 30+ critical bugs in MINRbase, a comprehensive database of nuclear- and mitochondrial-ribosomal-RNA-derived fragments (rRFs) using J2EE, JSP, and jQuery, facilitating its publication in the Nucleic Acids Research journal.
  • Focused on optimizing software development components of MINRbase, enhancing functionality and user engagement.
  • Executed a successful migration of the LicensePlates tool to Django, ensuring efficient RNA sequence analysis and dataset identification.
  • Contributed to the identification and classification of RNA sequences in LicensePlates, aiding in the recognition of miRNA/isomiR, tRF, and rRF.
  • Technologies: Java, Python, SQL, jQuery

Amazon Web Services

Software Development Engineer

  • Member of the AWS Supply Chain product team, specializing in the Demand Forecasting module.
  • Drive the backend development for demand forecasting and planning, ensuring system robustness, scalability, and efficiency.
  • Collaborate with cross-functional teams to integrate and manage data processes, without the direct ownership of ML models.
  • Influence architectural decisions, database management, and API design to ensure seamless data flow and optimal functionality.
  • Uphold AWS's standards of operational excellence by continuously integrating industry best practices and technological innovations.
  • Technologies: Kotlin, Java, TypeScript.

Education

Stony Brook University

August 2021 - December 2022

Masters in Computer Science

Coursework: Theory of Database Systems, Artificial Intelligence, Data Science Fundamentals, Computer Vision, Smart Energy in the Information Age.

University of Mumbai

August 2018 - July 2021

Bachelor of Engineering in Computer Engineering

Thakur Polytechnic

August 2015 - July 2018

Diploma in Computer Engineering

Publications

Social Media Intelligence for Brand Analysis

July 2021

IEEE International Conference On Computing, Communication and Networking Technologies (ICCCNT)

This paper aims to provide a real-time solution to stream the tweets from Twitter related to a specific brand and analyse them in real-time in order to extract insights from the tweet data that allows the brands to know what exactly are the social media users’ perceptions about them.

Paper Link

Code

Twitter Data Mining for Targeted Marketing

July 2020

IEEE International Conference on Inventive Research in Computing Applications (ICIRCA)

This paper aims to provide an approach to target users on Twitter by discerning their preferences with the use of unique keywords in the real estate sector.

Paper Link

Security System Based on Sclera Recognition

February 2018

International Research Journal of Engineering & Technology

The vessel patterns of Sclera are unique for each individual. Considering this uniqueness, a biometric system was developed to authenticate users based on their Sclera. This system captures a photo of the user's eye and preprocesses the image to extract vessel patterns. Further, the extracted vessel patterns are matched with the existing users' vessel patterns from the database to authenticate the user.

Paper Link

Code

Projects

Socialis

Developed a real-time social media intelligence tool to stream tweets from Twitter pertaining to a specific brand that performs various types of analyses based on the textual and location data and plots the insights graphically on a dashboard in real-time to enable the brand to better understand their audience with intelligent insights.

View Project

Twitter Profile Analyzer

Developed a Twitter analytics application that extracts rich insights from any public Twitter profile. Retrieved and analyzed data about tweets, likes, followers and location. Implemented techniques like sentiment analysis, named entity recognition and so on.

View Project

Security System Based on Sclera Recognition

The vessel patterns of Sclera are unique for each individual. Considering this uniqueness, a biometric system was developed to authenticate users based on their Sclera. This system captures a photo of the user's eye and preprocesses the image to extract vessel patterns. Further, the extracted vessel patterns are matched with the existing users' vessel patterns from the database to authenticate the user.

View Project

LinkedIn Followers Bot

Designed a Bot using Selenium to send connection requests in bulk to recommended profiles. Added a delay of 3-7 seconds to avoid getting detected by LinkedIn.

View Project

Blog Website

Launched an online blog website that allows users to create, update and delete posts. Implemented features such as user authentication session management, cloud image uploading and email verification.

View Project

Twitter Data Mining for Targeted Marketing

Engineered scripts to scrape data from Twitter about tweets, user profiles and location to target users on Twitter by discerning their preferences with the use of unique keywords in the real estate sector.

View Project

Skills and Technologies

Get in Touch