Principal Marketing Data Scientist

Key Skills

C# Programming, Database, Database testing, Design, Java Programming, Machine learning techniques, Natural Language Processing (NLP), Python Programming, R Programming, REST API, SQL, Text segmentation

Job Description

NielsenIQ BASES is a forward-thinking, rapidly growing, market research division of the NielsenIQ Company. That means we maintain an unbeatable client list, create best-in-class solutions and have access to incredible resources—without sacrificing the benefits of a smaller, leaner, close-knit company. We celebrate curiosity and creativity and encourage openness and collaboration.

We’re highly driven, team-oriented and psyched to use cutting-edge technologies to change how the world’s leading consumer companies innovate.

Through a blend of innovative market research technologies (such as evolutionary optimization, specialized choice modeling, neuroscience) and client consulting, we help major consumer products companies dramatically identify breakthrough product ideas, messaging that inspires action, fresh package designs and optimal launch strategies.

We work in a field that brings together human cognition, consumer behavior, statistical modeling, machine learning, and software development. Our work seeks to leverage NielsenIQ’s vast consumer and sales data assets, in combination with our industry-leading primary consumer research methods, to help our clients succeed with their innovations and new product launches. Our research and development efforts focus on developing more accurate and more scalable success prediction and volumetric forecasting models, to support the expanding scope of our business and our clients’ needs.

As a principal marketing data scientist, you will be leading or contributing to the research, design, testing, and implementation of various advanced marketing models. These may range from consumer choice and decision-making models to volumetric forecasting models, marketing mix models, and assortment and pricing optimization algorithms.

What you will do:

• Develop statistical and machine learning models to predict consumer behavior, integrating survey, household, and retail sales data
• Validate and calibrate current learning algorithms to estimate the success likelihood and volume potential of future products
• Develop models for estimating the impact of various advertising and marketing activities on sales
• Design and implement stochastic simulations for testing and validating various models and algorithms
• Prototype machine learning and optimization pipelines that integrate different data sources of different levels of granularity
• Keep up with the state-of-the-art in relevant areas – Statistics, Machine Learning, Operations Research, Marketing Science, Artificial Intelligence

 We are looking for people with

• Ph.D.(or Masters degree with relevant experience) in Marketing Science, Operations Research, Computer Science, Statistics or other relevant fields, with a high level of academic achievement
• Solid understanding of mathematical modeling, probability theory, and statistics, including Bayesian inference, and the design and simulation of stochastic systems
• Proven record of working with statistical learning algorithms such as Maximum Likelihood Estimation (MLE), Hierarchical Regression Models, Mixture Models, Hidden Markov Models (HMMs), and Markov Chain Monte-Carlo (MCMC) sampling
• Proven experience working with marketing models, including marketing mix models, forecasting models, and/or choice models.
• Firm knowledge of classical machine learning techniques such as random forests and boosting trees, SVMs, centroid-based and hierarchical clustering algorithms
• 4+ years of experience with scripting languages, in particular Python and R
• Strong knowledge of Relational Databases and SQL programming
• Excellent communication skills, written, oral and graphical, and the ability to present complex ideas in a clear and concise manner to a variety of audiences
• Experience with advanced experimental designs and adaptive sampling methods is preferred
• Knowledge of causal inference especially graph-based techniques is preferred
• Familiarity with Discrete/Combinatorial Optimization techniques
• Experience with text mining, NLP and NLU is preferred
• Experience with relational databases and SQL programming is preferred
• Experience working in cloud environments, especially Azure is preferred
• Experience with Web Services and REST APIs is preferred
• 2+ years of object-oriented programming experience (Java and C#) is preferred

Additional Information

Role can be opened anywhere in US - Can be remote or tied to any US office location.