Machine Learning Data Science

Programming Languages: Python, C, C++

Data Analytics & Visualization: Pandas, NumPy, Matplotlib, Seaborn, Excel, Power BI, Tableau Machine Learning & AI: Scikit-learn, TensorFlow, PyTorch Databases & Querying: RDBMS, SQL

Development Tools: Git, Jupyter Notebooks, Anaconda Soft Skills: Analytical Thinking, Data Storytelling, Communication, Team Collaboration, Time Management B-Tech Artificial Intelligence And Data Science AISSMS IOIT, Pune. CGPA:7.7 [2026] HSC H.P.T. ARTS & R.Y.K. SCIENCE College. 77.67% [2022] SSC Kalavati Nandalal Kela High School. 92.20% [2020] SKILLS

EDUCATION

ACADEMIC PROJECTS

Deep Fake Video Detection System. LINK

Smart US Visa Approval Classifier. LINK

Hybrid-Movie-Recommender-using-Cosine-Similarity LINK End-to-End Chest Cancer Classification using MLflow & DVC. LINK

+91-853******* ***************@*****.***

Built an novel model combining VGG16 and CNN, achieving 94% accuracy in detecting deepfakes, outperforming state-of-the-art techniques like NAS-Net.

Validated the model using confusion matrix analysis on a benchmark deepfake dataset, demonstrating high detection efficiency. Technologies used: Keras, VGG16, NAS-Net, Mobile-Net. Developed a machine learning model to predict H-1B visa approvals using historical data. Performed EDA and feature engineering to train classifiers (Logistic Regression, XGBoost) achieving over 88% accuracy. Containerized the application with Docker and deployed it on AWS EC2 via ECR, integrating GitHub Actions for CI/CD. Built an interactive Streamlit web app for real-time predictions. Built and deployed a deep learning model (92.5% accuracy) using TensorFlow. Implemented model deployment pipeline using AWS EC2 and Flask API, enabling real-time inference. Integrated MLflow for experiment tracking and DVC for dataset versioning, ensuring reproducibility and smooth collaboration. Automated CI/CD pipeline for seamless model training, evaluation, and cloud deployment. Implementation of a collaborative filtering-based system using user-based and item-based collaborative filtering algorithms 1,500+ unique user interactions for business intelligence. Evaluated the content based movie recommender system using Cosine similarity and Vectorization. CERTIFICATIONS

DOCUMENTATION SECRETARY.

AISA- Artificial Intelligence Students Association. DEPARTMENT CORDINATOR .

GATE QUALIFIED . Data Science and Artificial Intelligence. [2025]

[2023-25]

[2022-23]

IOIT-Association for Computing Machinery Student Chapter. HARSH GANGURDE

linkedin.com/in/harshgangurde github.com/HarshGangurde

ACHIEVEMENTS

EXTRA-CURRICULAR ACTIVITIES

Object Oriented Programming.

University of London & Goldsmiths

Object Oriented Programming Specialization.(Coursera). LINK

Microsoft Azure Administrator, AI Fundamentals LINK Generative AI.

Microsoft.

Microsoft Power BI Data Analyst. LINK

Microsoft.

Contact this candidate