What is Data Science and the Workflow

The career everyone is talking about

Data science has been called "the sexiest job of the 21st century", and the hype is backed by reality: it's one of the highest-paid and most in-demand fields in tech. But what is it? At its core, data science is the practice of turning raw data into useful insights and decisions. Every company today collects mountains of data — sales, users, clicks, sensors — and data scientists are the people who make sense of it, finding patterns, answering questions, and predicting what comes next.

The work spans a spectrum. A data analyst explores data to answer business questions ("which products sell best in which cities?"). A data scientist goes further, building models that predict the future ("which customers are likely to cancel?"). Both start from the same foundation — the skills you'll build in this course — and Python is the tool that powers it all.

Why Python for data science?

Python dominates data science for a simple reason: it has an incredible ecosystem of libraries built exactly for this work. You don't build everything from scratch — you stand on the shoulders of powerful, battle-tested tools:

• NumPy — fast numerical computing on arrays of numbers.
• pandas — handling tables of data (the workhorse you'll use most).
• Matplotlib and Seaborn — creating charts and visualizations.
• scikit-learn — machine learning models (the next step beyond this course).

Together these turn Python into a complete data science laboratory. This course teaches the first three deeply, giving you the foundation that everything else builds on. (If you've done our Python course, you're perfectly prepared; if not, a little Python basics will help you follow along.)

The data science workflow

Real data science follows a recognisable process. Understanding these stages gives you a map for every project you'll ever do:

1. Get the data — load it from files (CSV, Excel), databases, or the web.
2. Clean the data — fix missing values, wrong types, duplicates. This is often the biggest part of the job.
3. Explore and analyse — slice, group, and summarise to understand what the data contains.
4. Visualize — create charts that reveal patterns and tell the story clearly.
5. Model and predict — (the machine learning step) build models that forecast or classify.
6. Communicate — share the insights so people can act on them.

A surprising truth that every newcomer learns: stages 2 and 3 — cleaning and exploring — take up most of a data scientist's time. Real data is messy. The glamorous modelling is a small slice; the patient work of wrangling data into shape is the bulk. This course follows the workflow in order, so you'll build each skill exactly where it fits.

A taste of what's possible

To see where we're heading, here's a tiny preview using pandas — loading a data file and instantly summarising it. Don't worry about the details yet; just feel the power:

import pandas as pd

# Load a spreadsheet of data in one line
df = pd.read_csv("sales.csv")

# Instantly see the first few rows
print(df.head())

# Get summary statistics for every numeric column
print(df.describe())

# Answer a real question: average sales per region
print(df.groupby("region")["sales"].mean())

In four lines, we loaded data, previewed it, summarised it, and answered a business question. That's the leverage that makes data science so powerful — and what you'll be able to do yourself by the end of this course. Tools called Jupyter notebooks are the usual home for this work (they let you run code in small interactive chunks alongside notes and charts), but the playground here lets you start immediately. Next, we'll begin with NumPy, the foundation of fast numerical computing in Python.