Junior Data Scientist Interview Questions UK: 8 Real Questions and What Interviewers Want

Applying for a Junior Data Scientist job in the UK? Here is what you are likely to be asked. These questions come from our Junior Data Scientist question bank of 16 questions, written for UK employers at junior level.

Eight Junior Data Scientist interview questions, and what the interviewer is thinking

1. What do you think a graduate data scientist actually does day to day, and why this career?

Why they ask: They want to check your expectations are realistic, because much of the job is data preparation rather than modelling.

2. Talk me through the stages of a typical data science project, from the business question to deployment. Which stage usually takes longest?

Why they ask: Understanding the full lifecycle shows you can work beyond a single notebook.

3. How would you handle categorical variables and feature scaling before training a model, and when does each matter?

Why they ask: These preprocessing choices are everyday tasks and affect model results.

4. What are train, validation and test sets, and how does cross-validation work? How would you adapt this for time series data?

Why they ask: Correct validation underpins trustworthy results and is a very common junior test.

5. How does a random forest work, and how does it differ from gradient boosting? When might you choose one over the other?

Why they ask: Tree ensembles are the workhorse models on tabular business data.

6. How would you evaluate a regression model, and what do metrics like RMSE, MAE and R-squared tell you?

Why they ask: Metric choice and interpretation are fundamentals the interviewer will test.

7. Your manager gives you a vague question, 'Why are our sales falling?', and a messy dataset. How do you begin?

Why they ask: Junior data scientists often receive ambiguous briefs, and clarifying them is an essential skill.

8. Tell me about a data science or analytics project you are proud of, from your degree, a placement or personal work.

Why they ask: As a graduate, your projects are your evidence of skill, and they want to hear your decisions as well as results.

A sample answer to question 1

What do you think a graduate data scientist actually does day to day, and why this career?

From what I have learnt and from speaking to practitioners, a large share of the day goes on understanding the business question, finding and cleaning data, and exploring it, with modelling being a smaller slice than many graduates expect. A junior would typically work on a defined part of a larger project: writing SQL and Python to build datasets, creating features, trying baseline models in scikit-learn, evaluating them, and documenting results in notebooks or short presentations for stakeholders. There is also code review, version control with Git, and attending stand-ups. I chose this career because I enjoy turning a vague question into evidence someone can act on. During my degree dissertation I predicted student retention from admissions data, and the most rewarding part was not the algorithm but discovering which factors mattered and explaining them to staff who could use that. I also like that the field keeps changing, so there is always something new to learn, and that good work can make a visible difference to real decisions.

Common mistake to avoid: Expecting to build deep learning models all day; Ignoring communication

Want the rest?

Full model answers, key points and common mistakes for all 16 questions are on the Junior Data Scientist interview questions page. The first five are free, no sign-up needed.