Data Science
Turn data into answers.
Data scientists find patterns, run experiments and build models that guide business decisions.
What the work looks like
Exploring data, building models, running A/B tests, presenting insights to stakeholders.
Key skills
- Statistics & probability
- Python (pandas, scikit-learn)
- SQL
- Experiment design
- Storytelling with data
Common tools
Jupyterpandasscikit-learnTableau / Power BISQL
Pros
- Mix of maths, coding and business
- Visible impact on decisions
- Many industries
Cons
- Entry-level roles are competitive
- Much time spent cleaning data
- Role definitions vary a lot
A 12-month starter roadmap
Months 1–3
- • Statistics basics
- • Python + pandas
- • SQL
- • Data visualisation
Months 4–6
- • Machine learning fundamentals
- • Kaggle competitions
- • A/B testing
- • 2 analysis projects
Months 7–12
- • Advanced ML
- • Business case studies
- • Portfolio blog
- • Internship applications
Reality check
Students like: Discovering something nobody knew.
Students dislike: 80% data cleaning.
Hardest part: Communicating uncertainty to non-technical people.
Common myth: It's not mostly deep learning — it's mostly statistics and SQL.