[nom]

Aspiring Data Scientist | Statistical Modeling | ML | NLP | Python & R

[téléphone] • [e-mail] • LinkedIn • GitHub • Kerala, India • Open to Remote

PROFESSIONAL SUMMARY

Aspiring Data Scientist with a deep academic foundation in Statistics and Machine Learning, currently pursuing a PG

Diploma in Data Science from the University of Calicut. Skilled in Python, R, SQL, and end-to-end data science workflows — including statistical inference, predictive modeling, NLP, and data-driven storytelling. Deployed a live NLP model (Streamlit) and executed multiple ML and statistical projects. Seeking a remote Data Scientist internship or junior role to solve real-world problems through rigorous data analysis and machine learning.

EDUCATION

PG Diploma in Data Science | University of Calicut · 2025 – Present

Relevant: Machine Learning, Statistical Inference, Data Visualization, Python for Data Science

B.Sc. Statistics with Data Science | University of Calicut · 2021 – 2024

CGPA: 6.975 / 10 | 69.75% (First Class)

Key Coursework: Probability Distributions, Statistical Inference, Regression Analysis, Sampling Theory, Multivariate

Techniques, Time Series Analysis, Statistical Quality Control, Linear Algebra & Numerical Methods, Python Programming, DBMS, Data Visualization (Python & Tableau), Introduction to ML

TECHNICAL SKILLS

Languages · Python, R, SQL

Statistics · Hypothesis Testing, Confidence Intervals, Time Series, Multivariate Analysis,

Regression Diagnostics, Sampling Theory

ML & NLP · Regression, Classification, Clustering, TF-IDF, Naive Bayes, Model Evaluation,

Feature Engineering

Data Analysis · Pandas, NumPy, EDA, Data Cleaning, Feature Engineering, Excel

Visualization · Matplotlib, Seaborn, Tableau (basic), Power BI (basic)

Tools & Dev · Git, GitHub, Jupyter Notebook, VS Code, Streamlit, LaTeX

Database · SQL Queries, Joins, Aggregations, DBMS Concepts

Soft Skills · Analytical Thinking, Statistical Reasoning, Problem Solving, Communication

EXPERIENCE

Data Science Practitioner (Project-Based) | Self-Directed Learning · 2024 – Present

• Built and deployed 3 live Streamlit ML applications: SMS spam classifier, Telco churn predictor, and Premier League match predictor — all publicly accessible via Streamlit Cloud.
• Implemented XGBoost models for classification (churn prediction, sports outcome prediction) with class imbalance handling, custom probability thresholds, and ROC-AUC evaluation.
• Engineered features across multiple projects: lag features, time-based variables, UK public holiday indicators, team performance metrics, and TF-IDF text vectorization.
• Built an end-to-end retail demand forecasting pipeline on real UK sales data — data cleaning, feature engineering, regression modeling, and business-friendly stock recommendation output.
• Maintained all projects on GitHub with structured READMEs, requirements.txt files, and serialized model artifacts (Pickle) for full reproducibility and cloud deployment.

PROJECTS

▸ SMS Spam Classifier — Deployed NLP App

End-to-end NLP pipeline: text preprocessing, TF-IDF vectorization, Naive Bayes classifier. Serialized model with Pickle and deployed as a live Streamlit web app. Achieved high precision and recall on SMS spam dataset. | Live:

[lien] | GitHub : [lien]

▸ Telco Customer Churn Prediction — Deployed ML App

XGBoost classifier predicting telecom customer churn. Handled class imbalance using scale_pos_weight; applied custom threshold (0.30) to improve recall; evaluated with ROC-AUC. Deployed live on Streamlit Cloud with interactive sliders and real-time probability score display. | Live: [lien] | GitHub:

[lien]

▸ Premier League Match Predictor — Sports Analytics App

ML system predicting Premier League outcomes (home win/draw/away win) using XGBoost on 10 years of match data

(2014-2025). Integrated Monte Carlo simulation for scoreline probabilities and Poisson goal modeling. Dashboard shows win probability bars, expected goals, and head-to-head history. | Live: premier-league-match-predictor-

[lien] | GitHub : [lien]

▸ Smart Inventory Recommendation System — Retail ML App

End-to-end demand forecasting pipeline on real UK retail data. Built lag features, time-based variables, and UK public holiday indicators. Regression model outputs product-level demand ranges with safety buffer stock recommendations via an interactive Streamlit dashboard. | GitHub : [lien]

ADDITIONAL INFORMATION

• Actively building a data science portfolio with deployed projects on GitHub ([lien]).
• Continuous learner — deepening expertise in advanced ML, deep learning fundamentals, and cloud-based data tools.
• Preparing for data science entrance assessments; strengthening competitive programming and quantitative reasoning skills.
• Collaborated in academic teams for data analysis projects and technical presentations.
• Actively seeking remote Data Scientist internships and junior roles globally — open to freelance data projects.