About me

I am a Master's student in Data Science at Indiana University–Bloomington, with a Bachelor's degree in Information Technology from Dwarkadas J. Sanghvi College of Engineering in Mumbai, India. I am enthusiastic about applying machine learning and data-driven solutions to real-world problems, combining analytical rigor and problem-solving skills.

As an ML intern at AMD, I optimized image preprocessing pipelines to enhance real-time lane detection performance. At Swan Solutions and Services, I developed an ensemble model for aspect-based customer segmentation, enabling more targeted business improvements. My notable projects include creating quantized models for NSFW classification, which achieved faster inference speeds and more efficient model architectures. I also designed an end-to-end ML pipeline for a network security dataset, employing MLOps tools like MongoDB, MLFlow, DagsHub, GitHub CI/CD, Docker, AWS S3, ECR, and EC2. Additionally, I implemented cataract detection using Explainable AI for improved transparency and interpretability and developed a privacy-preserving federated learning framework for medical diagnosis to ensure robust data security. Other key projects include a speech emotion recognition system, a Natural Language SQL query generator for an e-commerce database powered by GCS and Gemini 1.5, and a GCP-based analysis of IPL data. Proficient in machine learning and deep learning frameworks, as well as statistical methods, I am actively advancing my skills in MLOps. My skill set encompasses programming languages, databases, and essential software and cloud tools for data analysis and model development, enabling me to effectively tackle complex problems.

What i'm doing

  • design icon

    Grad Student

    Pursuing a MSc in Data Science from Indiana University-Bloomington.

  • Web development icon

    Currently Learning...

    Learning and working with MLOps tools to build end-to-end ML pipelines.

Resume

Education

  1. Indiana University-Bloomington

    Aug 2023 — May 2025

    Master of Science in Data Science

    GPA: 3.92/4.0

  2. Dwarkadas J. Sanghvi College of Engineering

    Aug 2019 — May 2023

    Bachelor of Technology in Information Technology

    GPA: 9.01/10.0

Experience

  1. Indiana University–Bloomington

    Machine Learning Intern

    May 2024 — Aug 2024
    • Developed non-invasive elephant recognition methods using optimized VGG-19 architectures to detect elephants based on distinctive wrinkle patterns, resulting in a 20% improvement over baseline accuracy.
    • Implemented SIFT for feature matching of complex wrinkle patterns in elephant images, resulting in an increase of 15% accuracy.
  2. AMD (Advanced Micro Devices)

    Machine Learning Intern

    July 2022 — Oct 2022
    • Devised an image preprocessing pipeline using PyOpenCL, integrating weighted-average and Gaussian Blur, followed by Canny Edge Detector, Region of Interest selection, and Hough transformation for real-time lane detection.
    • Accelerated pipeline processing by 12% with PyOpenCL on AMD hardware, outperforming CPU-only execution.
  3. Swan Solutions and Services

    Data Science Intern

    July 2021 — Oct 2021
    • Developed an ensemble model combining LDA for aspect extraction, TextBlob for sentiment identification, and a Logistic Regression ML model with 96.6% peak accuracy to facilitate aspect-based customer segmentation.
    • Conducted root cause analysis utilizing LDA to identify critical improvement areas, reducing customer complaints by 20% and increasing positive reviews by 15%.

My skills

    Overview

    Machine Learning, Deep Learning, Explainable AI, NLP, Computer Vision, Data Analysis, Statistical Analysis

    Programming Languages

    Python, R (Statistical Analysis), C, Java, JavaScript, SQL

  • Python
    90%
  • SQL
    80%

    Frameworks & Tools

    PyTorch, TensorFlow, Keras, scikit-learn, PySyft, OpenCV, NLTK, AWS, GCP, Spark, Docker, MLFlow, DagsHub, Databricks, Tableau, Power BI, Looker, Git, GiHub Actions, Streamlit

  • scikit-learn
    90%
  • PyTorch
    90%
  • NLTK
    75%

    Databases

    Firebase, Microsoft SQL Server, MySQL, PostgreSQL, SQLite, BigQuery, MongoDB, Neo4j

    (* "%" denotes the most frequently used , not proficiency)

  • PostgreSQL
    50%
  • MySQL
    30%

    Libraries

    PyTorch, TensorFlow, Keras, scikit-learn, NLTK, PySyft, OpenCV, PyOpenCL, Pillow, Librosa, Gensim, TextBlob, langdetect, multiprocessing, NumPy, Pandas, Matplotlib, Plotly, Streamlit

Projects & Publications

Contact

Contact Form