My learning began with numbers — structured, logical, certain. But it’s in data, with all its imperfections and hidden stories, that I found real curiosity. Over the last 2+ years, I’ve immersed myself in building machine learning and deep learning projects — not as assignments, but as explorations of how machines learn, how patterns emerge, and how real-world problems can be reframed through algorithms. From behavioral classification to image recognition, each project has shaped not just my skills but the way I think. I now seek a data science internship where I can contribute with depth, continue to question what I build, and grow in an environment that values both thinking and doing.
View My WorkI didn’t enter this field with a plan — I entered it with questions. Studying Mathematics at Ramanujan College, DU, and now pursuing my M.Sc. in Mathematics with Computer Science from Jamia Millia Islamia, I’ve learned to respect the structure of theory — but I’ve also learned to challenge it. While others stopped at solving equations, I started turning data into stories, models into tools, and curiosity into code. For the past 2+ years, I’ve been building machine learning and deep learning projects — not because someone asked me to, but because I couldn’t not do it. From personality prediction models to deep learning image classifiers, every project taught me something books couldn’t: how messy real-world data is, how algorithms fail before they work, and how insight often lives where you least expect it. I’m not here to just “do data science.” I’m here to learn like it matters, build like it solves something, and grow with people who care about both the why and the how. If you're working on something honest, hard, or a little bit chaotic — I’d like to be involved.
I have experience with Python, Java, C, SQL, HTML, and a wide array of libraries and tools including Pandas, NumPy, Scikit-learn, Matplotlib, Seaborn, XGBoost, TensorFlow, and Keras. My projects demonstrate a strong ability in data preprocessing, feature engineering, and delivering high-fidelity predictive models.
Python, Java, C, SQL, HTML
Object-Oriented Programming (OOP), Data Structures, Algorithms
Missing Value Imputation, Feature Engineering, Categorical Encoding (Label, One-Hot), Feature Scaling, Text Feature Extraction.
Statistical Analysis, Data Visualization (Matplotlib, Seaborn), Correlation Analysis.
MATLAB, LaTeX, MS Office (Word, Excel, PowerPoint), Overleaf, Jupyter notebook, Python, Pandas, NumPy, Scikit-learn, Matplotlib, Seaborn, XGBoost, Tensorflow, Keras.
Matplotlib, Seaborn, PlotLy.
Developed a robust classification model predicting student social media addiction, leveraging advanced feature engineering and comprehensive EDA. Validated top-performing algorithms (Random Forest, KNN) to 92% cross-validation accuracy, revealing key behavioral drivers of addiction.
Learn More →Developed a CNN-based image classification model for Nike, Adidas, and Converse products. Implemented transfer learning VGG16 and utilized Keras callbacks for robust model training and performance optimization.
Learn More →Architected a Netflix content classification system by extracting high-value features from unstructured text and metadata. Achieved top-tier classification accuracy across diverse ML models (XGBoost, Random Forest).
Learn More →Developed a high-fidelity classification model predicting personality traits (Extrovert/Introvert) from behavioral data. Delivered leading performance via SVM and MLP Classifier (92% F1 score and 93% Recall through K-Fold CV).Deployed the model using FastAPI to enable real-time personality prediction via REST API.
Learn More →Have a project in mind or just want to say hello? Feel free to reach out via email or connect with me on LinkedIn and GitHub.
Email: sanyalpoushali19@gmail.com