Data Scientist

We work with political parties, investors, media organisations, think tanks, NGOs and companies of all sizes around the world to help them understand public opinion, how it affects them, and what they should do in response to it. 

We have developed the most accurate models for predicting election outcomes that exist in the market, and have deployed them in the UK, US, Canada, Australia, France and elsewhere. 

As we expand, we are building a dedicated data science function to refine and optimise our existing models, and to create new ones that help us to maintain our position as a leader in the field of public opinion.

We’re looking for someone who’s comfortable working with unstructured data, and can turn it into clear outputs that deliver actionable insight for clients. You’ll be curious about public opinion and how a richer understanding of it can be of value to clients. We welcome applicants at different career stages; the title, responsibilities and compensation will reflect your background.

Salary: Competitive, and dependent on experience

Contract: Full time and permanent

Location: London

What you’ll be doing:

Your days will be varied and full, with plenty of opportunity to learn. You’ll be:

• Building and deploying MRP (and other) models to help our clients understand likely election outcomes
• Understanding the range of our client’s problems and creatively designing technical solutions to help solve them
• Managing data pipelines, designing quantitative inputs to models, assuring data quality, designing post-stratification frames
• Mining data for insight, exploring public and private data creatively with a view to unearthing new ways of thinking about a client problem
• Visualising outputs and explaining them to clients, prospects and stakeholders

What you’ll bring:

• Proficiency with core analysis tools, at minimum: Excel, plus R or Python (and you can work independently, not just with AI-assisted coding).
• Strong skills in:
• Data wrangling/manipulation (e.g., dplyr, pandas)
• Visualisation (e.g., ggplot2, matplotlib)
• Statistical modelling (e.g., multilevel models, Bayesian methods, machine learning)
• Experience working hands-on with real-world datasets: cleaning, validating, and interpreting results (including uncertainty and statistical significance).
• Ability to turn analysis into a clear narrative, producing client-ready outputs with effective charts and explanations.
• A strong interest in electoral politics, political science and public opinion.
• Confident communication skills and a strong client service ethic.

Nice to have:

• Familiarity with MRP concepts; ideally experience building multilevel regression models in practice.
• Experience working with survey and public opinion data, and/or government/census datasets.
• Experience with (or interest in) interactive outputs (dashboards, web apps, etc.).
• Working knowledge of reproducible workflows (e.g., Git/version control, reusable code, basic data engineering or data architecture principles).

How to apply

Email your CV and a cover letter to by 30 January 2026 with the subject line: Data Science Role.

Your cover letter should include:

1. Why you’re applying, and how you think you can help Stack succeed.

2. In no more than two paragraphs, your answer to the following:

• How do you view the role of AI in data science and data analysis?

3. In no more than three paragraphs, your response to the following scenario:

• Pick any political party in a democratic nation, preferably one that is not currently leading in public polls. Imagine that you have been commissioned by the party to run a large sample poll of their country, with the goal of producing a one-page briefing to direct decision making. Outline briefly (at most 1 paragraph) the challenges faced by the party, then outline (in the remaining space) how you would approach this research, which should include what you would collect in the poll, the analyses you would expect to run on the results, and how you would use those results and that analysis to advise the party. 

These exercises are designed to test your own thinking and writing, so please make sure that your answers reflect original work. We understand that tools such as ChatGPT can be useful in the writing process but we encourage you to avoid submitting answers which are fully AI-generated. Submissions that rely too heavily on LLMs/AI are often easy to recognise and will negatively affect your application.

Stack Data Strategy is an equal opportunities employer and we welcome applications from all suitably qualified people regardless of your age, any disability you may have, your gender or gender reassignment, marital or civil partner status, pregnancy or maternity, race, religion or belief, colour, nationality, ethnic or national origin, sex or sexual orientation. As part of our commitment to equal opportunities, we are always open to discussions about alternative working patterns or hours and try to maximise the scope for talented people to contribute to Stack’s mission!

Please let us know if you need any reasonable adjustments to be made during recruitment or in employment for reasons associated with your physical or mental health. We are committed to ensuring everyone is able to contribute to Stack in whatever way is most appropriate for them.