Learning Analytics Data Scientist

Job summary

Tribal’s vision is to empower the world of education and we make it our mission to provide the software and services which underpin student success. The mission of the Product Management and Product Development Department is to support the empowerment of the world of Education through the definition of a market-shaping product strategy, delivered to a market-leading standard of quality and value The Product Management Team supports this mission through possessing deep market insight and clear understanding of Education market opportunities which drives a market leading, innovative Product and solution strategy. The role of the Data Scientist is to develop data integration/transformation pipelines, machine learning models and predictive analytics methods where performance can be automatically optimised and applied to live customer datasets and which are used within the product platform to enable customers to understand and optimise student success.

Job details

• To extract and work with data from multiple large structured and unstructured sources, including streaming log data pertaining to student activities with teaching and learning resources and academic and progression data extracted from student management systems, and transform data in high performance data pipelines.
• To work closely with the product development and professional services teams to process datasets from disparate sources and manipulate raw data to ensure that it is appropriate for analysis.
• To research application of machine learning methods to solve customer problems in the education sector
• To work closely with customers to understand their student success challenges and translate these to analytics-focused solutions that provide customers with insight to help them overcome these problems, combining a thorough understanding of the education domain and its data, with data science techniques and technical knowledge about how to apply machine learning methods and algorithms.
• To develop new data products which are embedded in Tribal’s customer facing products and services.
• To analyse customer datasets to identify important variables correlated with student outcomes and derive new indicators of student risk and success.
• To work directly with customer and public datasets to combine disparate datasets and develop predictive models using machine learning techniques.
• To analyse time series and log data and perform feature engineering to extract predictive features.
• To develop, enhance and maintain the predictive analytics engine · To evaluate machine learning methods and techniques for new data sources.
• To design and build solutions which scale to large amounts of data, particularly relevant for xAPI student activity data.
• To develop techniques and prototypes for processing, analysing and visualising student, learning and activity data
• To develop and enhance the data analytics infrastructure used for internal data analysis and data science R&D
• To represent Tribal at events/conferences and promote Tribal as a thought leader in Learning Analytics, including speaking at events, writing white papers and blog posts, and fostering networks/relationships in the evolving Learning Analytics sector.
• To present key findings from data analysis to non-technical audiences, inspiring them by putting data into context through the telling of stories with their data.
• To contribute to Tribal’s data science and analytics strategy for embedding analytics within Tribal’s products and services.
• To influence the strategy and direction of the platform based on an understanding of data science, the education domain and its data.
• To support the marketing and sales efforts, including contribution to tender responses, providing information for marketing material, carrying out sales demonstrations and presentations to potential customers.

Role deliverables

• A thorough understanding of the market and its challenges means that analytics focused solutions can be provided to customer problems.
• Both static and updating sources of data are processed, analysed, manipulated and transformed into high performance data pipelines.
• The data pipeline consistently updates outputs and deals gracefully with unexpected data from source institutions.
• Model outputs and development techniques are presented in a way which is clearly understood by both technical and non-technical audiences.
• Tribal’s data analytics infrastructure is continually enhanced and data analytics embedded into Tribal’s products and services.
• Standards are evolving for data models in the education sectors worldwide, and for customer institutions.
• A strong contribution is evident to the development of and work within a robust and repeatable procedure for data model development

Qualifications, skills, knowledge and experience

Qualifications – Essential

• A strong technical or software development background.
• Postgraduate degree (to Masters or PhD level) with a strong analytical or computational focus (Computer Science, Mathematics, Physics, Statistics, Data Science etc.).

Qualifications – Desirable

• Proven track record with data analysis in the education sector.

Skills and Abilities – Essential

• Good interpersonal skills including strong verbal and written communication.
• Strong presentation skills.
• Able to work in a multidisciplinary environment and be a good team player.
• Meticulous attention to detail, accurate and thorough; delivering high quality output to tight deadlines.
• Ability to prioritise, communicate and deliver in the face of complex projects and time-sensitive deliverables.
• Great problem solving and analytical skills combined with the ability to explain concepts to both technical and non-technical audiences.
• Ability to use Python to build data pipelines and machine learning solutions, and intimately familiar with Python’s core data science libraries including pandas, numpy, matplotlib, scikit-learn and seaborn.
• Abilty to evaluate model performance and understand how to build models against imbalanced dataset.
• Aptitude for quickly learning new technical skills and becoming proficient quickly.

Skills and Abilities – Desirable

• Creative and forward thinking – able to come up with innovative analytics focused solutions to solve new problems.

Knowledge and Experience – Essential

• Working knowledge of design methods and principles and ability to deploy methods appropriately.
• Substantial experience of working with statistics and data analysis using statistical and machine learning techniques.
• Substantial experience of data mining and machine learning.
• Experience of the further and higher education sector and its standards for data interoperability, such as HESA or AVETMISS.
• Experience of managing analytics projects using CRISP-DM or a similar project management methodology.
• Experience in the interpretation and review of specifications.
• Experience in writing and amending software, building commercial software solutions and developing enterprise solutions using object orientated methods.
• Experience in writing efficient T-SQL and Stored Procedures.
• Strong working knowledge of database platforms including SQL Server (or equivalent) and MongoDB.
• Strong working knowledge of Linux configuration and scripting (e.g. bash on Ubuntu).
• Experience in building and interfacing with web APIs using REST and SOAP.
• Strong working knowledge of version control using Git.
• Experience in feature engineering and data transformation.
• Practical experience of the methods used to apply machine learning algorithms to complex datasets.
• Experience of building data processing pipelines which integrate data from multiple sources.
• Experience of ETL and a solid understanding of database design.

Knowledge and Experience – Desirable

• Knowledge of R.
• Experience in deploying apps using Docker.
• Familiarity with machine learning ensemble methods applied to data fusion.
• Experience of data visualisation using d3.js and JavaScript or similar.
• Experience of working with xAPI.
• Experience of using Apache Spark to build data pipelines and machine learning solutions against large structured and unstructured datasets.