Open to opportunities · Bangalore, India

Hey, I'm
Vignesh Kumar

Data enthusiast & software developer building intelligent systems — from ML pipelines to full-stack apps. I turn messy data into decisions that matter.

Machine LearningData SciencePython SQLWeb DevelopmentLogistic Regression Churn PredictionLinear RegressionEDA GANsStyle TransferIoT BlockchainCloudGenAILLMsBangalore Machine LearningData SciencePython SQLWeb DevelopmentLogistic Regression Churn PredictionLinear RegressionEDA GANsStyle TransferIoT BlockchainCloudGenAILLMsBangalore

Who I am

I'm Vignesh Kumar, a developer and data scientist based in Bangalore, India. I enjoy solving real-world problems with data — building everything from predictive models to location-aware applications.

My work spans machine learning, data analysis, and web development. I'm especially drawn to projects where a well-designed model or a crisp dataset insight can change how decisions are made.

When I'm not coding, I'm exploring new ideas, writing notes, or tinkering with tools that make workflows faster. This site is my corner of the internet — a living home for projects, ideas, blogs, and experiments.

Python Machine Learning SQL Data Analysis HTML/JS Jupyter scikit-learn Pandas
13
Public Repositories
7+
ML Case Studies
Ideas in Progress
BLR
Based in Bangalore

Things I've built

A collection of projects ranging from data science case studies to full-stack apps. All source code lives on GitHub.

🚕

A location-based taxi aggregator that compares fares and availability across providers in real time. Built with HTML and geolocation APIs to help riders pick the best option instantly.

HTML
📊

Logistic regression model that scores sales leads and predicts conversion probability. Helped a fictional EdTech firm prioritise outreach, pushing conversion rates toward a 80% target.

Jupyter Notebook
📡

End-to-end churn prediction pipeline using Random Forest, XGBoost, and other classification algorithms. Identifies high-value customers most likely to leave, enabling proactive retention strategies.

Jupyter Notebook
🎬

SQL-driven analysis of a movie production company's data to recommend the best genres, directors, and actors for their next global release. Pure SQL, zero fluff — just insights from data.

SQL
🚲

Linear regression model predicting bike-sharing demand based on weather, season, and time-of-day signals. Helps rental operators optimize fleet allocation and reduce idle capacity.

Jupyter Notebook
💳

Deep exploratory data analysis of credit application data to surface risk patterns and default indicators. Uncovers correlations, distributions, and outliers that drive lending decisions.

Jupyter Notebook
🎨

Generative Adversarial Network that transplants the artistic style of one image onto the content of another. A deep-learning exploration of image synthesis and neural creativity.

Jupyter Notebook
☁️

A collection of hands-on mini-projects spanning Cloud computing, IoT sensor pipelines, and Blockchain fundamentals — experiments at the intersection of three emerging tech domains.

Python
🏥

A Generative AI–powered assistant tailored for medical research — uses LLMs to interpret clinical queries, surface relevant findings, and deliver accurate, context-aware responses to support healthcare decision-making.

Python
View all 13 repos on GitHub →

Writing & thoughts

Notes on machine learning, software craft, and things I learn along the way. Posts coming soon!

📝 Logistic Regression Demystified A practical walkthrough of lead scoring with real data. coming soon
🌀 Surviving Churn: ML for Telecom How ensemble models beat vanilla classifiers on imbalanced data. coming soon
🗺️ Building Location-Aware Apps Lessons from building the taxi aggregator with browser APIs. coming soon

Ideas & experiments

A public scratchpad — concepts I'm exploring, half-baked projects, and questions I want to answer.

🤖 Auto-EDA Tool Drag-and-drop dataset → instant exploratory analysis report. exploring
🗃️ SQL Query Explainer Paste a SQL query, get a plain-English breakdown with a query plan. exploring
📈 Personal Finance Tracker Minimal dashboard to track spending, built with local-first data. exploring

Tools I use

My everyday stack — the languages, libraries, and platforms I reach for first.

🐍 Python + scikit-learn Go-to for all ML projects — models, pipelines, evaluation.
🐼 Pandas + NumPy Data wrangling, cleaning, and numerical crunching.
🛢️ SQL Analytics queries, joins, window functions — the classics.
📓 Jupyter Notebooks Exploratory analysis, visualisation, and storytelling with data.
🌐 HTML / CSS / JS Frontend for quick tools, dashboards, and personal projects.
🐙 Git + GitHub Version control, collaboration, and portfolio hosting.