Data Scientist

Position Overview

The LRNG Data Scientist utilizes a knowledge of statistics, machine learning, and data mining tools to realize the potential of games and digital learning experiences for engagement, learning and assessment. The person in this role collaborates with clients, internal teams and external partners to apply and create efficient, robust approaches to data extraction and reporting, file cleaning, feature engineering and model building while ensuring the accuracy, utility and accessibility of results and interpretations. Lastly, the LRNG Data Scientist provides support, when needed, to identify and develop external partnerships with leading figures in the fields of learning analytics, educational data mining and intelligent systems for purposes of joint projects that advance the work in the field of digital learning and assessment.

The LRNG Data Scientist is expected to take on the following roles and responsibilities:

Primary Duties and Responsibilities

Data Flow and Analysis

• Collaborate with external partners and internal teams to improve and maintain a complete data pipeline from initial data generation within digital learning experiences through multiple stages of cleaning, parsing, analyzing, reporting and archiving data.
• Create well phrased queries in SQL to pull down data from the LRNG servers, in order to clean, parse and annotate telemetry.
• Design and implement descriptive reports summarising patterns in the use of the LRNG platform.
• Apply data mining and learning analytic techniques, and more traditional statistical models where needed, to investigate, identify and confirm relevant patterns of behavior within games and digital learning experiences in order to support inferences about learners’ cognitive and affective states.
• Support success of internal content and third party learning content, integrating knowledge of predictive modeling to make data-driven recommendations for improving digital learning and assessment tools.
• Lead the use of predictive modeling to support iteration and improvement of games and digital learning experiences for learning and assessment and play a key role in the success of the organization’s services to external developers.

Research and Partnership Development

• Contribute to advancement of the organization’s research agenda.
• Initiate and cultivate communications and partnerships with leading experts in the areas of learning analytics, educational data mining and intelligent systems with the aim of growing the organization’s capacity for developing and applying high leverage approaches to making generalizable inferences about learner abilities and other states.
• Investigate and evaluate novel approaches to working with telemetry from game-based and interactive digital experiences in order to incorporate the most promising approaches and grow the organization’s skill base.
• Act as a ‘voice’ for the organization’s analytics, presenting the team’s approaches to data mining and analytics to external audiences.
• Contribute to fundraising efforts as needed.

Minimum Qualifications

• Strong expertise in all stages of predictive model development including authorship and implementation of database queries, feature engineering, model development, model selection and model maintenance.
• Competency with SQL and data analysis tools such as MySQL, Hadoop, NumPy, Pandas the R programming language, MatLab and/or data mining packages (e.g. RapidMiner and WEKA).
• Experience successfully handling complex telemetry data, including developing expedient processes for cleaning and aggregating data.
• Ability to effectively communicate complex concepts to a lay audience.
• Ability to collaborate effectively with a diverse and geographically distributed team.
• Comfortable working in fast-paced dynamic environment focused on product development.
• 2+ years of successful experience using SQL and predictive modeling techniques.

Preferred Qualifications

The ideal candidate will also have the following preferred prior experience, skills and dispositions:

• Experience working with multi-level and educational data.
• Knowledge of and/or a proven interest in Natural Language Processing and approaches to processing and automated scoring of text with tools such as LightSIDE.
• Exposure and/or experience developing and applying deep learning techniques.
• Strong verbal and written communication skills.

Education

Required: Master’s degree or equivalent in data sciences, statistics or related field.

Application

Resumes or CVs should be sent to jobs@collectiveshift.org.