Collaborated and assisted multiple teams in requirements analysis, decision making
I had understood the client's needs and responded according to the project requirements by conducting requirements analysis, data collection and analysis, assisted in decision making, etc.
Automated data movement tasks using Apache Airflow DAGs (Directed Acyclic Graphs), enabling scheduled, reliable, and repeatable data transfers for insurance-related processes.
Orchestrated and managed data movement workflows in Apache Airflow to facilitate seamless extraction, ingestion, and movement of data across diverse sources and destinations within the project ecosystem.
Graduate Assistant
University Of Maryland, Baltimore County, UMBC
02.2022 - 05.2023
Maintained detailed student and class records, including project and participation grades.
Assessed student progress with course material using fair and equitable grading policies.
Boosted student learning through explorative discussions and group activities.
Intern
Orbit Shifters (Start-Up)
12.2020 - 05.2021
6 Months online internship on developing AI products for Gaming Industry using Computer Vision
Involved in annotating the images, supporting in model building and testing the model
Performed Data Pre-processing activities, creating ML/DL Models.
Developed an Android app in Android Studio using Java and Google Maps API to increase successful issue resolution by 90% for drivers in need of vehicle breakdown assistance.
Hosted Firebase Cloud Messaging for communication between drivers and assistance providers.
Ecommerce Website
Deployed an eCommerce website using Django, Python, Bootstrap, CSS, and client-side JavaScript, achieving 100% uptime.
Connected PayPal as a payment method and designed a user-friendly checkout flow for both registered and guest users, achieving a high conversion rate of 20%.
Improved Response time by 15% by resolving bugs to optimize web APIs and provide a smooth user experience.
Food Ordering System
The project includes the development of login and logout functionality and is built using a microservice architecture with Eureka Service Registry and Elastic search-log stash to generate log files.
The backend application is developed in Java and Spring Boot, while the frontend uses HTML, JavaScript, and Bootstrap.
Amazon Sentiment Review Analysis using PySpark
Conducted sentiment analysis on 400k Amazon reviews using PySpark and a dataset sourced from Kaggle (700MB).
Employed data preprocessing techniques like tokenization, stop word removal, and CountVectorizer to convert text into a bag of words and utilized a Random Forest classifier to predict sentiment with 76% accuracy.
Evaluated model performance through cross-validation, gaining valuable insights from the results.
Demonstrated proficiency in PySpark, NLP, and machine learning to process and analyze large-scale data effectively.