About me

I'm a USC Master's Grad Student. I love to build advanced ML algorithms to solve impactful business problems. I’m overall a person who likes to take up challenges which are intellectually stimulating and over the past few years I have honed my skills through various projects.

I am passionate about using data-driven insights to drive positive change and solve real-world challenges. If you are looking for a Data Scientist/Machine Learning Engineer with a track record of delivering innovative solutions, I would love to connect and discuss how I can contribute to your organization.

What i'm doing

  • analytics icon

    Data Analytics

    Unveiling insights through the artful interpretation of data, crafting actionable intelligence from information's intricate patterns.

  • Pred Model icon

    Predictive Modeling

    Decoding the future with smart algorithms, turning patterns into predictions for informed decision-making.

  • stats icon

    Statistics

    Unveiling truths through rigorous hypothesis testing, translating complex data into actionable insights with precision.

  • data science icon

    Data Science

    Crafting actionable intelligence from data insights, decoding the future with smart algorithms, and unveiling truths through rigorous statistics.

Resume (Download PDF)

Education

  1. University of Southern California

    Aug 2021 — May 2023

    M.Sc., Machine Learning and Data Science GPA 3.81/4.0
    Relevant Courses: Machine Learning, Deep Learning, Databases, Applied and Cloud Computing, Data Structure and Algorithms, Linear Algebra, Probability Theory, Digital Signal Processing

  2. Vellore Institute of Technology

    July 2017 — July 2021

    B.Tech., Electronics and Communication Engineering GPA 8.95/10.0
    Relevant Courses: Machine Learning, Data Mining and Predictive Analysis, Computer Vision, Big Data Analytics

Experience

  1. Data Scientist @ Prime Healthcare

    Feb 2024 — Present

    ◦ Leveraged Prophet (Meta Open Source) to analyze supply utilization and purchasing trends, optimizing reorder points and order quantities. Reduced stockouts by 30%, saving $1.2 million annually in expired inventory and capital.
    ◦ Developed a RAG model with Llama 3.1 to web scrape and categorize healthcare items, suggesting substitutes to enhance purchasing decisions. Improved item substitution accuracy by 25%, streamlining procurement and reducing manual categorization efforts.

  2. Machine Learning Intern @ CarmaCam

    Aug 2023 — Feb 2024

    Devising two approaches to identify and classify road signs for autonomous vehicles:
    (1) AutoML on Google Cloud platform, and
    (2) transfer learning with various architectures (ResNet50, Xception, and InceptionResNetV2).

  3. Data Scientist @ USC ITS - Office of CISO

    Feb 2022 — May 2023

    ◦ Redesigned the risk prediction framework, achieving improved F1-score of 0.91 for 28,000 vendors of USC.
    ◦ Implemented XGBoost model, accomplished 15% reduction of false positives, through rigorous A/B testing.
    ◦ Automated processes for alerting vendors of their risk ratings on Power BI, provided data analysis findings to stakeholders with recommendations to mitigate vendor risks. Cut down 20+ hours of weekly manual work.

  4. Data Science Research Intern @ Vellore Institute of Technology

    Nov 2020 — July 2021

    ◦ Implemented novel efficient deep-learning model to diagnose patients with COVID-19 or pneumonia from X-ray images.
    ◦ Employed Unet encoder-decoder models, improved training speeds by a factor of 2, achieving low FLOPs comparable to state-of-the art models.
    ◦ Deployed this network achieving 99.3% accuracy and 99.31% F1-score in Micronet M3 model.

  5. Machine Learning Intern @ Arista Networks

    Nov 2020 — July 2021

    ◦ Received theoretical as well as hands-on training on concepts of fingerprinting along with ML algorithms in 1 week.
    ◦ Leveraged k-Nearest Neighbor and Random Forest models to estimate user position in an indoor environment. Using Wi-Fi and inertial sensors yielded positioning as precise as 2-3 m.
    ◦ Designed algorithm to apply concepts of RSSI to extract real-time location of client devices operating on access points of WiFi routers placed across work facility with an accuracy of 0.98.

My skills

  • ML Modeling
    90%
  • Data Analysis
    80%
  • Coding Languages (Python, SQL)
    90%
  • Python Libraries - scikit-learn, PyTorch, TensorFlow, Pandas, NumPy, Matplotlib, OpenCV
    90%
  • Tools (AWS, GCP, PowerBI, Anaconda, GitHub)
    80%

Contact

Contact Form