👨💻 Professional Summary
Master's Student in Data Science with 1 year of research and practical experience in Data Science and Machine Learning. Demonstrated proficiency in Python, Java, SQL, statistical modeling, and Keras (LSTM/RNN), Scikit-learn, Data Visualization, etc. Strong history of projects and research contributions, including an LLM chatbot and predictive modeling applications, displaying advanced problem-solving abilities and technical expertise. Driven to use innovative machine learning and data science techniques to solve challenging problems in the Artificial Intelligence industry and apply academic research to create effective, data-informed solutions.
🎓 Education
Coursework:
Tools & Technologies:
Models Used:
Core Engineering Coursework:
💼 Professional Experience
- Contributed to projects including Fraud Detection, Lung Cancer Prediction, Vehicle Price Prediction, and Animal Classification, applying data science techniques to solve real-world problems in healthcare, finance, and automotive domains.
- Designed and implemented ML models and algorithms with data preprocessing, feature engineering, and hyperparameter tuning, improving prediction accuracy and model efficiency across projects.
- Collaborated with cross-functional teams to analyze large datasets, delivering actionable insights and robust model evaluation metrics to ensure reliable outcomes.
- Utilized Python, scikit-learn, and related libraries to develop predictive models and workflows for each project.
- Presented results and visualizations to stakeholders, guiding data-driven decisions and demonstrating model impact on project objectives.
🚀 Data & AI Projects Portfolio
- Utilized AWS S3 buckets for secure data storage and delivery, integrating with web interface to enable seamless data access and management.
- Created and configured an EC2 instance on Ubuntu to host a dynamic web application, demonstrating proficiency in cloud deployment and server management.
- Developed a Retrieval-Augmented Generation (RAG) system using OpenAI GPT models to extract and compare structured specifications from unstructured PDF documents.
- Built an LLM-powered comparison pipeline with PDF parsing, semantic extraction, Streamlit visualization, and Excel export, ensuring robust handling of missing data.
- Built a Generative AI-based chatbot using machine learning and NLP to answer design exam-related queries accurately.
- Integrated SYL blog content to deliver context-aware responses, study strategies, and step-by-step exam preparation guidance.
- Enhanced user engagement and preparation efficiency by providing personalized tips, ideas, and solution-focused insights.
- Built comparative forecasting models (LSTM, RNN, ARIMA) for COVID-19 daily confirmed cases.
- Performed stationarity testing (ADF), seasonal decomposition, and moving average smoothing.
- Achieved superior performance using LSTM for nonlinear pandemic wave modeling.
- Evaluated models using RMSE and R² metrics.
- Built an end-to-end credit card fraud detection system using Python, Scikit-learn, and SMOTE, engineering behavioral and time-based risk features from large-scale transaction data.
- Improved fraud detection performance with Stratified Cross-Validation (ROC-AUC: 0.976) and performed model evaluation using confusion matrix, precision-recall curves, and feature importance analysis.
- Developed a 15-class image classification system using a custom CNN and MobileNetV2 transfer learning, training on ~1,900+ images with data augmentation.
- Achieved 73.8% validation accuracy using transfer learning, significantly outperforming the baseline CNN (6% accuracy).
- Implemented early stopping, learning rate scheduling, confusion matrix analysis, and classification reports to optimize model performance; deployed inference pipeline and saved trained models for reuse.
- Developed a regression-based vehicle price prediction system using Random Forest and Ridge models, achieving R² = 0.79 on test data.
- Built an end-to-end ML pipeline with data cleaning, feature engineering (vehicle age, log transformation), OneHotEncoding, and StandardScaler using scikit-learn Pipelines.
- Implemented model evaluation (MSE, R², residual analysis) and deployed a reusable prediction function with model persistence using Pickle.
- Developed a web application for stock trend prediction and automated trading using predictive analytics and machine learning.
- Implemented blockchain-based security to ensure secure transactions and robust financial data protection.
- Designed an intuitive, visually appealing UI to deliver a seamless user experience for secure economic management.
- Developed a time-series forecasting model using Facebook Prophet on 17+ years of market price data.
- Performed data preprocessing, feature engineering, and seasonal trend analysis.
- Forecasted future vegetable prices with confidence intervals for 365 days.
- Visualized trend and seasonality components for market insight.
📚 Professional Certifications & Memberships
-
Machine Learning Engineer Learning Path2025 - Present
Google Cloud / Google Skills -
Complete Guide to SQL for Data Engineering: from Beginner to AdvancedFeb 2026
LinkedIn Learning / Deepak Goyal -
SQL Tips and Tricks for Data ScienceFeb 2026
LinkedIn Learning -
Data Science Foundations: Data EngineeringFeb 2026
LinkedIn Learning -
The Complete Python Bootcamp from Zero to HeroJan 2025
Udemy -
JS and Full Stack DevelopmentDec 2024
Udemy -
Java - The Complete Java Development BootcampApr 2024
Udemy -
Machine Learning from TeachnookSep 2022
Teachnook