## TL;DR
Data science combines statistics, computer science, and domain expertise to extract insights from data. The modern workflow — from data wrangling through machine learning to deployment — is supported by Python's ecosystem and cloud infrastructure.

## Core Explanation
The data science lifecycle: problem formulation→data acquisition→cleaning→EDA→feature engineering→modeling→evaluation→deployment→monitoring. Data quality is the dominant determinant of project success — "garbage in, garbage out."

## Detailed Analysis
Statistical foundations: probability distributions, hypothesis testing, confidence intervals. ML paradigms: supervised (classification, regression), unsupervised (clustering, dimensionality reduction), reinforcement learning. Model evaluation must match business objectives.

## Further Reading
- Kaggle: Data Science Competitions
- Towards Data Science
- Journal of Data Science