Lead Research Data Scientist | Fraud Detection in Market Research

Location: Reading, King’s Road

We’re the world’s leading data, insights, and consulting company; we shape the brands of tomorrow by better understanding people everywhere.

About the job

We’re seeking an experienced, passionate Lead Research Data Scientist to spearhead our research-driven initiatives in detecting and preventing fraud and poor data quality within online market research surveys. From bot responses to panel profile manipulation, you’ll develop and validate advanced models and systems that uphold the integrity of our data and the robustness of our insights.

This is a high-impact role at the intersection of behavioural analytics, machine learning, research methodology, and the latest advancements in LLMs and AI, ideal for someone passionate about applying rigorous scientific methods in a global industry context.

Job Goals

Design, validate, and deploy novel machine learning models to detect fraudulent survey responses, including bots, duplicate entries, low-quality survey data and improbable or non-relevant responses.

Develop and maintain real-time scoring systems in production to assess respondent authenticity and engagement.

Conduct in-depth analysis of behavioural patterns, metadata, and response timing to uncover anomalies and suspicious activity.

Collaborate with survey operations, panel management, and research teams to embed fraud detection tools into survey workflows.

Patent novel algorithms & approaches to fraud detection within Market Research.

Provide thought leadership and mentorship to the team; promote best practices in ethical data handling, reproducible research, and experimental design.

Contribute to the wider Kantar data science ecosystem by presenting methodologies, publishing internal white papers, and facilitating knowledge exchange on fraud detection.

Stay abreast of academic and industry developments in fraud tactics, data validation, and respondent quality assurance.

Ideal Skills & Capabilities

PhD in Data Science (or a highly relevant MSc plus real-world research experience), Statistics, Mathematical Modelling, Computer Science, or a related quantitative field.

Seniority and proven experience in data science, with at least 2 years focused on fraud detection, survey analytics, automated data quality solutions, or related research applications.

Strong proficiency in Python, R, SQL, and machine learning libraries (e.g., scikit-learn, XGBoost, TensorFlow, pyTorch). Experience with Kafka, ML Ops, and CI/CD pipelines is advantageous.

Experience with ML Ops & cloud platforms such as Azure ML and AWS is required.

Deep understanding of anomaly detection, behavioural modelling, and time-series analysis.

Deep research experience with NLP techniques for validating open-ended responses including applications Deep Learning for Gen AI / LLMs. Experience with evaluation of LLMs such as LLM as a judge & human evaluated testing.

Strong communication skills, with the ability to translate technical findings into research insights and actionable recommendations.

What We Offer

• Competitive salary and performance-based bonuses
• Flexible working hours and hybrid work model
• Access to rich global survey datasets and cutting-edge tools, including state-of-the-art fraud detection models
• A collaborative, mission-driven team focused on data integrity, reproducibility, and innovation
• Opportunities for professional development, academic collaboration, and leadership growth
• Support for presenting at academic and industry conferences, and contributing to peer-reviewed publications where appropriate

Kantar is committed to inclusion and diversity; therefore, we welcome applications from all sections of society and do not discriminate based on age, race, religion, gender, pregnancy, sexual orientation, gender identity, disability, marital status, or any other legally protected characteristics.