B.Tech in Artificial Intelligence | Pune, India 🇮🇳

I'm a data analyst focused on building end-to-end analytical solutions — from raw data ingestion and SQL modeling to machine learning pipelines and interactive dashboards. My work sits at the intersection of data engineering, analytics, and applied ML.

• 🎓 Education: G H Raisoni Institute of Engineering & Business Management, Jalgaon
• 💼 Experience: Data Science Intern @ 3RI Technologies Pvt Ltd (Jan 2024 – Jun 2024)
• 🔍 Currently: Open to Data Analyst roles (Entry–Mid level)
• 🏗️ Latest Project: EV Charging Station Demand Analysis — 23-country ML pipeline (Python + MySQL + Prophet + Power BI + Streamlit)
• 📊 Tools I reach for first: Python, SQL, Power BI, Pandas, Scikit-learn
• 🧠 100+ LeetCode problems solved (SQL + Python)
• 📬 Reach me: sumersingpatil0193@gmail.com

From data extraction and analysis to machine learning and deployment, these are the technologies powering my projects.

Python

SQL

HTML5

CSS3

NumPy

Pandas

Scikit-learn

TensorFlow

Matplotlib

Streamlit

Power BI

Jupyter

VS Code

Git

Multi-country infrastructure analysis across 23 European nations — end-to-end ML + BI pipeline

• Cleaned and engineered synthetic dirty data; modeled in MySQL with analytical SQL views
• Built Random Forest demand classifier + Prophet time-series forecasting
• Deployed interactive Streamlit dashboard with Folium geospatial maps, Plotly charts, and 9 analytical tabs
• Full pipeline: Raw CSV → MySQL → Jupyter (EDA + ML) → Streamlit → Power BI

Stack: Python MySQL Streamlit Scikit-learn Prophet Folium Plotly Power BI

NLP-powered classification system on 17,880 job postings — AUC-ROC: 0.984

• Built end-to-end ML pipeline: TF-IDF vectorization → Logistic Regression → SMOTE for class imbalance
• Added SHAP explainability to surface top features driving fake job predictions
• Integrated MySQL (6 analytical views), Power BI dashboard, and deployed on Streamlit Cloud
• Chi-square hypothesis testing added to validate statistical significance of findings

Stack: Python NLP Scikit-learn MySQL Power BI Streamlit SHAP SMOTE

Content-based filtering system using cosine similarity + TMDB API integration

• Recommends movies based on genre, cast, director, and keywords
• Fetches live movie posters, trailers, and metadata via TMDB API
• Deployed as a Streamlit web app with clean UI

Stack: Python Streamlit Scikit-learn Pandas TMDB API

EDA on 1,000+ student records — identifying performance trends across demographics

• Cleaned and explored dataset using Python (Pandas, Seaborn) to uncover trends across gender, parental education, and test preparation
• Executed MySQL queries to identify key factors influencing Math, Reading, and Writing scores
• Created visualizations using Matplotlib & Seaborn to present score distributions and demographic patterns

Stack: Python MySQL Pandas Seaborn Matplotlib