Problem framing
Defines the user, decision, baseline, constraints, error costs, and evidence before modelling.
Selects a sophisticated method before clarifying what result would be useful.
Software, Product, Data & AI
Data Scientist Job Description
A Data Scientist uses statistical reasoning, experimentation, and modelling to answer consequential questions or build data-informed product capabilities.
Define the decisions, product context, data maturity, deployment support, and scientific depth instead of starting with an algorithm list.
Candidate-facing template
Edit the description for your organisation
Replace bracketed details, remove anything that is not genuinely required, and obtain the appropriate internal approval before advertising.
Ready-to-edit job description
Copying excludes all recruiter notes below.
Data Scientist
- Location
- [Location or working arrangement]
- Employment type
- [Full-time or contract]
- Reports to
- [Data Science Manager or Head of Data]
About the role
[Company name] is hiring a Data Scientist to work on [decision or product area]. You will turn ambiguous questions into testable approaches, prepare and evaluate data, develop suitable analyses or models, communicate uncertainty, and help teams use results responsibly.
What you will be responsible for
- Frame decisions and hypotheses with stakeholders before choosing a method.
- Explore, clean, link, and document data while testing quality, bias, leakage, and suitability.
- Develop and compare statistical, experimental, causal, forecasting, or machine-learning approaches.
- Validate results against meaningful baselines and explain uncertainty, limits, and likely failure conditions.
- Partner on deployment, monitoring, governance, communication, and post-launch evaluation where models enter use.
What success looks like
- Data-science work changes a defined decision, product behaviour, or validated understanding.
- Models outperform appropriate baselines on measures connected to real use.
- Stakeholders understand uncertainty, affected groups, limitations, and monitoring needs.
Essential qualifications
- Applied experience with statistics, experimentation, modelling, and data preparation.
- Strong SQL plus a suitable analytical language such as Python or R.
- Ability to validate models and communicate assumptions to technical and non-technical audiences.
- Responsible judgement around privacy, fairness, security, interpretation, and model impact.
Preferred qualifications
- Advanced study or equivalent practical depth in a quantitative discipline relevant to the work.
- Experience deploying, monitoring, experimenting on, or governing models in the target domain.
Tools and working knowledge
- SQL, notebooks, and analytical programming libraries
- Version control, experiment tracking, data, and model platforms
- Visualisation, documentation, orchestration, and monitoring tools
Compensation: [Add approved range, currency, bonus or equity, benefits, compute resources, and location basis.]
[Company name] will discuss reasonable adjustments and offer accessible alternatives for live coding, whiteboards, or timed exercises.
How to apply
Apply through [method] with a project explaining the decision, data, method, contribution, validation, limitations, and outcome.
Performance expectations
What good performance looks like
Use these outcomes to replace vague activity lists with the evidence the hiring manager expects to see after the person joins.
Projects begin with a decision, baseline, user, cost of error, and evidence threshold.
Validation reflects likely production or decision conditions rather than only an offline metric.
Unsuccessful experiments and null findings are documented and inform the next choice.
Hiring-manager intake
Questions to settle before advertising
Record specific answers so sourcing, screening, and interview decisions use the same definition of the role.
- 1
Which decisions, models, experiments, or product behaviours will the scientist own?
- 2
What data exists, who governs it, and which quality or access constraints are known?
- 3
Who supports engineering, experimentation, deployment, monitoring, and domain validation?
- 4
What statistical depth is essential versus familiar tooling that can be learned?
Evaluation criteria
Evidence to use in a Data Scientist scorecard
Agree the criteria before interviews begin, then score examples against the same evidence standard.
Validation judgement
Tests leakage, bias, uncertainty, robustness, baselines, and real-use conditions.
Reports one high metric without explaining data construction or failure modes.
Decision communication
Explains what is known, uncertain, actionable, and unsafe to infer.
Uses technical authority to overstate causality or confidence.
Structured interview
Data Scientist interview questions and strong signals
Ask the same core questions in the same order, then use follow-ups to understand the candidate’s individual contribution.
1
Tell me about a model that did not outperform a simple baseline.
Strong answer signal
Shows honest validation, diagnosis, decision value, and learning rather than metric hunting.
2
How would you choose the error metric for a high-impact classification decision?
Strong answer signal
Connects affected groups, base rates, false outcomes, thresholds, operations, and monitoring.
3
Describe a stakeholder question you refused to answer from the available data.
Strong answer signal
Identifies the inference risk, communicates it clearly, and proposes safer evidence or action.
Related titles
Check the scope behind the title
Decision Scientist — often focuses on experimentation, causal questions, and business decisions.
Applied Scientist — may carry deeper research or algorithm development in a product setting.
Machine Learning Scientist — usually concentrates on model research rather than production engineering.
Common hiring mistakes
Problems to remove before publishing
Combining analytics, data engineering, ML engineering, research, and BI ownership in one vacancy.
Requiring a doctorate where applied evidence would demonstrate the needed depth.
Naming algorithms and cloud tools without describing the decision, data, baseline, or deployment reality.
Continue the hiring workflow
Related roles and recruiter resources
Related job descriptions
Helpful recruiter resources
AI Candidate Screening Software
Structure candidate evidence while retaining human review.
Candidate Database Management
Maintain searchable, current candidate skills and context.
Résumé Score Checker
Compare a résumé with explicit role criteria.
AI Candidate Matching
Review opportunities and limits in data-assisted matching.
Source and review notes
How this template was prepared
- Language
- International English
- Prepared by
- ATZ CRM Editorial Team
- Review
- ATZ CRM Recruitment Editorial Review
- Last reviewed
- 2026-08-05
O*NET: Data Scientists
Reference for occupation tasks, knowledge, skills, abilities, and work activities.
ESCO: occupations and skills
Reference for internationally recognised occupation and skills terminology.
Qualifications, compensation, licences, working conditions, and equal-opportunity wording must be checked for the role and location before use.
Recruiter questions
Data Scientist job description FAQs
What should a Data Scientist job description include?
Describe the decision or product, data, methods, validation, deployment support, responsible-use expectations, collaboration, and outcomes.
How is a Data Scientist different from a Data Analyst?
Data Scientists more often build statistical or machine-learning models and experiments. Data Analysts commonly focus on reporting, investigation, and decision support.
Is a doctorate required?
Only for work needing that research depth. Applied projects, industry experience, postgraduate study, or equivalent evidence may suit many roles.