📍 MA, USA | 📧 [email protected] | 📞 857-972-3001

• Check out more of my projects on GitHub!
• Check out my learning blog here - Data Wisdom: My Learning Journey of Data & Technology

I am a seasoned data professional. Over the course of 5 years, I've worn multiple hats - Data Scientist, Data Engineer, Data Integration Specialist, and Data Analyst, with a profound background in Machine Learning and Data Science. My expertise spans across Python, SQL, Machine Learning, ETL tools, Tableau, SAP HANA, BI tools and much more. I take pride in delivering data solutions that drive decision-making and offer insights.

• Data Science: Expertise in Exploratory Data Analysis, Feature Engineering, A/B testing, Predictive Analytics, and Statistical Modeling.
• Machine Learning: Proficient in frameworks such as TensorFlow, Keras, and Scikit-learn, with hands-on experience in model selection, optimization, and deployment.
• Natural Language Processing: Successfully implemented sentiment analysis and multi-label text classification projects.
• Data Engineering: Strong background in ETL processes, Data Migration, and Database Warehousing using tools like SAP BO Data Services and SAP HANA.
• Data Visualization: Created insightful visualizations using Tableau, Looker, and Power BI, among others.

• Data Analyst / Data Scientist, Internship, University of Massachusetts President’s Office, MA, Dec 2022 - Dec 2023
  Led initiatives for machine learning models, improved decision-making, and operational expenses reduction.
• Data Engineer, Tata Consultancy Services (Client – ARM), Jun 2020 - Dec 2021
  Designed data integration solutions, maintained data pipelines, and optimized data models in SAP HANA.
• Data Integration Specialist & Data Analyst, Tata Consultancy Services (Client – Ericsson), Feb 2017 – June 2020
  Excelled in data extraction, transformation, and loading; implemented sophisticated SQL queries and data visualizations.

Linkedin Profile

• Master’s in Computer Science, University of Massachusetts Lowell | GPA: 3.9/4
• Bachelor of Technology, Jawaharlal Nehru Technological University | GPA: 73.15%

• Pneumonia Detection Using Neural Networks: "Pneumonia Detection Using Neural Networks" is a pivotal project that utilizes the power of deep learning to identify the presence of pneumonia from chest X-ray images. The initiative leverages Convolutional Neural Networks (CNNs), Multi Layer Perceptron (MLP) renowned for their efficacy in image analysis tasks. The model was trained on a robust dataset comprising both normal and pneumonia-affected X-ray scans. Through a series of convolutional layers, pooling, and dense layers, the neural network effectively learned the distinguishing features of pneumonia in the scans. With the aid of data augmentation and fine-tuning, the model achieved impressive accuracy, paving the way for potential applications in rapid diagnostic tools and providing healthcare professionals with a valuable second opinion. This project underscores the potential of AI in revolutionizing medical diagnostics and the importance of accurate and early detection in improving patient outcomes.
• Movie Recommender system This is an advanced project focused on harnessing the expansive and rich dataset provided by TMDb to curate personalized movie recommendations for users. Tapping into the wealth of movie metadata, user ratings, and contextual information, the project employs sophisticated algorithms to deliver both content-based and collaborative filtering recommendations. By analyzing movie attributes such as genres, actors, directors, and plot summaries, the system offers content-based suggestions that align with a user's expressed preferences. Simultaneously, by examining user rating patterns, it identifies similarities between users to provide collaborative recommendations.
• Sentiment Analysis of Movie Reviews: This project delves into the realm of Natural Language Processing (NLP) to gauge the emotions and sentiments expressed in movie critiques. Utilizing a rich dataset of movie reviews from IMDb, the project employs advanced machine learning algorithms to classify reviews as positive, negative, or neutral. By processing textual data, tokenizing, and vectorizing reviews, the model effectively captures the essence and nuances of each review. Several techniques, including word embeddings, were experimented with to enhance the model's accuracy. The end result is an intelligent system capable of automatically determining the sentiment of a given movie review. This project not only showcases the power of NLP in understanding human emotions in textual data but also offers valuable insights for the entertainment industry to gauge audience reception of movies.
• Analysis and Defense Strategies against Adversarial Attacks on Convolutional Neural Networks This project is a comprehensive study into the vulnerabilities of Convolutional Neural Networks (CNNs) when faced with maliciously crafted adversarial inputs. Recognizing the potency of adversarial examples that can deceive a well-trained CNN, this project delves deep into their underlying mechanisms, highlighting potential entry points for such attacks. By examining various adversarial attack methods, the project establishes a framework to simulate these threats and evaluate a CNN's robustness. Moreover, it introduces a range of defense strategies, such as input preprocessing, adversarial training, and architectural modifications, to fortify CNNs against these attacks.
• Multi-Class Multi-Label Text Classification: This project is an advanced machine learning endeavor aimed at categorizing textual data into multiple predefined categories, with the potential for each piece of text to belong to multiple categories simultaneously. Drawing from a comprehensive dataset, this project involved preprocessing and vectorizing text to prepare it for model training. Leveraging techniques such as word embeddings and deep learning architectures, the model is trained to recognize patterns that indicate multiple labels. This project demonstrates the depth of classification tasks in NLP and the importance of accommodating the multifaceted nature of textual information.
• Chest X-Ray Classification to Detect Covid-19 Using Deep Neural Networks: "Chest X-Ray Classification to Detect Covid-19 Using Deep Neural Networks" is a cutting-edge project aimed at harnessing the capabilities of deep learning to assist in the early detection of Covid-19. Utilizing a curated dataset of chest X-ray images, both from Covid-19 patients and controls, the project employs three neural networks - LeNet, ResNet, VGG19 – specialized in image recognition tasks.
• Language Translator Telegram Bot This is an innovative project that seamlessly integrates the robust capabilities of the Microsoft Translator API into a user-friendly Telegram bot. This bot is designed to provide real-time language translations directly within the popular messaging platform, Telegram. Users can effortlessly send text messages in their native tongue, and the bot instantly responds with accurate translations in the desired target language.

Check out my projects

Feel free to connect with me for collaborations, discussions, or opportunities! 💬