Data Scientist
Data science is part statistics, part programming, part storytelling. Far more time goes into cleaning data and framing the right question than into building clever models.
- Work environment
- Office or remote, computer-based, embedded in product, finance or operations teams
- Typical route in
- Degree in a numerate subject typical; many enter via masters, apprenticeships or from software or analyst roles
- Progression
- Analyst or junior data scientist → data scientist → senior or lead → principal scientist, head of data, or into machine-learning engineering.
What you'll experience
- Statistics
- Programming
- Ambiguity
- Visualisation
- Presentation
- Experimentation
- Messy data
You might enjoy this if you…
- Enjoy maths and coding in combination
- Like open-ended questions without obvious answers
- Can explain complex ideas simply
- Are comfortable being wrong and iterating
- Get satisfaction from evidence changing a decision
Things people sometimes underestimate
- Most time goes on cleaning data, not building models
- The maths matters less than asking the right question
- Stakeholders often want a simple answer, not a sophisticated one
- A lot of work never ships — experiments get shelved
- Communication skills matter as much as technical ones
Worth thinking about
- Work can be ambiguous — you often define the problem yourself
- Results are probabilistic, which some audiences find frustrating
- Keeping skills current requires continuous learning
- Impact depends heavily on whether the organisation acts on your findings
Typical responsibilities
- Framing business questions as testable problems
- Cleaning and exploring large, messy datasets
- Building and validating statistical or machine-learning models
- Presenting findings to non-technical colleagues
- Monitoring models in use and retraining them
- Working with engineers to put analysis into production
Day in the life of a data scientist: common questions
What does a typical day in the life of a data scientist look like?
Data science is part statistics, part programming, part storytelling. Far more time goes into cleaning data and framing the right question than into building clever models. Day to day, the work includes framing business questions as testable problems, cleaning and exploring large, messy datasets, building and validating statistical or machine-learning models.
What qualifications do you need to become a data scientist?
Degree in a numerate subject typical; many enter via masters, apprenticeships or from software or analyst roles
What is the work environment like for a data scientist?
Office or remote, computer-based, embedded in product, finance or operations teams
What should you think about before choosing this career?
Work can be ambiguous — you often define the problem yourself Results are probabilistic, which some audiences find frustrating Keeping skills current requires continuous learning Impact depends heavily on whether the organisation acts on your findings
How does a data scientist progress in their career?
Analyst or junior data scientist → data scientist → senior or lead → principal scientist, head of data, or into machine-learning engineering.
Careers similar to data scientist
Students who explore this role often walk a day in these too.
- Day in the life of a game developerBuild the systems behind games — movement, combat, tools and performance — in a team where everything is judged on how it feels.
- Day in the life of a software developerBuild products, solve technical problems and work with teams to turn ideas into software.
- Day in the life of a accountantAnalyse financial information, ensure accuracy and compliance, and help individuals or businesses make sound decisions.
Every workplace is different. This experience is designed to give a realistic flavour of the role, not to predict exactly what your future job will be like.