I'm a Master’s student in Data Science at TU Braunschweig, passionate about Machine Learning, AI, and Data Science. I specialize in AI model development, predictive analytics, and data-driven decision-making. My expertise lies in deep learning, natural language processing, and advanced machine learning algorithms.

📍 Currently based in Braunschweig, Germany
🎯 Open to Data Science, Machine Learning, and AI Engineering roles

✅ Machine Learning & Deep Learning
✅ Natural Language Processing (NLP)
✅ Computer Vision
✅ Predictive Analytics & Time Series
✅ Cloud Computing
✅ Data Visualization (Matplotlib, Seaborn, Tableau)

🔹 Developed an LSTM-based NLP model to detect deceptive content
🔹 Implemented Logistic Regression as a baseline model for comparison
🔹 Integrated with an Anvil-based web app for real-time predictions
🔹 Applied text preprocessing, sentiment analysis, and deep learning techniques
🔗 GitHub Repo

🔹 Developed a Convolutional Neural Network (CNN) to classify brain MRI images into four categories: Glioma, Meningioma, Pituitary, and No Tumor
🔹 Achieved 96% test accuracy and weighted F1-score of 0.96
🔹 Utilized TensorFlow/Keras for model development and training
🔹 Applied data preprocessing, image augmentation, and early stopping to improve model performance
🔹 Deployed the model for predicting tumor types from unseen MRI images
🔗 GitHub Repo

🔹 Developed a predictive model using Random Forest & XGBoost to detect potential customer churn
🔹 Processed Telco Customer Churn dataset by handling missing values & encoding categorical features
🔹 Applied SMOTE to balance dataset & improve classification performance
🔹 Conducted Exploratory Data Analysis (EDA) to uncover key churn indicators
🔹 Optimized model performance via hyperparameter tuning (GridSearchCV, RandomizedSearchCV)
🔹 Evaluated models using Accuracy, Precision, Recall, F1-Score & Confusion Matrix
🔹 Deployed trained models using Pickle for real-time churn prediction
🔗 GitHub Repo

🔹 Developed a hybrid stock price forecasting model using LSTM & ARIMA
🔹 Collected SAP SE (SAP.DE) stock data from Yahoo Finance for analysis
🔹 Engineered sliding window features for LSTM-based deep learning predictions
🔹 Applied Auto ARIMA for optimal parameter selection in time-series forecasting
🔹 Compared model performance using MAE, RMSE, MAPE, SMAPE, and R²
🔹 Visualized trends using Matplotlib & Seaborn for better market insights
🔹 Achieved high accuracy (R²: 0.996 for LSTM, 0.997 for ARIMA)
🔗 GitHub Repo

🔹 Applied KMeans clustering to segment customers based on purchasing behavior
🔹 Engineered features like Recency, Frequency, and Monetary Value for analysis
🔹 Identified key customer segments: Retain, Re-Engage, and Nurture
🔹 Optimized cluster count using Elbow Method & Silhouette Score
🔹 Visualized insights using 3D scatter plots & violin plots for better interpretation
🔹 Used Python, Pandas, Scikit-learn, Matplotlib, and Seaborn
🔗 GitHub Repo

🔹 Developed a monthly dashboard tracking hospital metrics and patient data
🔹 Analyzed payer-wise revenue and cost breakdowns
🔹 Used Power BI for visualization and Excel for preprocessing
🔹 Enabled data-driven insights for regional demand and cost-saving opportunities
🔗 GitHub Repo

🏆 Knight Rank on LeetCode (Algorithmic Problem-Solving)
🏆 2nd Place in KICCS-D-HACK Coding Competition
🏆 Exceptional Performance Award @ Info Edge India

📜 Relevant Certifications:

• Machine Learning Specialization (Andrew Ng - Coursera)
  • Supervised Machine Learning: Regression and Classification
  • Advanced Learning Algorithms
  • Unsupervised Learning, Recommenders, Reinforcement Learning
• Deep Learning Specialization (Andrew Ng - Coursera)
• AI For Everyone (Coursera)
• Python for Data Science & AI (IBM - Coursera)

⭐ Explore my projects and feel free to connect! Always open to learning and collaborating on AI & Data Science projects. 🚀