It has been several years since a 2012 issue of the Harvard Business Review called data science the “sexiest job of the 21st century,'' yet attitudes towards the discipline remain as bullish as ever. One of the key reasons data science continues to be lauded as the world’s best career is the numerous career paths that it offers. “Data Science” is not a specific job, but a discipline that provides its practitioners with a variety of professional options.
The following are just a few of the jobs available to individuals with a data science degree.
Career #1: Data Science Generalist (Analyst)
As the “foundational” data science role, generalist data scientists focus on discovering useful insights within data. Whether they be small datasets or “big data”, generalist data scientists use experimental and exploratory methods to discover new insights, simplify complexities and create informed predictions about future events.
Data scientists in this role often work for industry, NGOs and government. Their work is often experimental and focuses on discovering unknown insights in data; however, they must also possess the problem solving abilities necessary to answer specific queries. Experienced generalists possess strong communication skills, as they need to be able to communicate their insights to non-data scientist decision-makers.
Data science generalists possess the foundational skills necessary to understand the work done by those in most specialised data science disciplines. Many data scientists that work in specialised fields begin their careers as generalists and learn to specialise through practical experience in a specific niche.
Career #2: Research Scientist (Analyst)
The job of a research data scientist is similar to the work of any other research scientist. Whether they work in academia, government or industry, research data scientists perform experiments with data in order to discover new insights into specific research topics.
Many research data scientists work in bioscience; for example, biostatisticians—data scientists that describe biological processes as statistical functions—use data science to test the effects of drugs on the human body without the need for lab equipment or tissue samples. This process is called “drug discovery”, and is used by both large research institutions and tiny startups.
The website BioRxiv features many examples of data science-driven research, such as this paper, which describes the use of computer vision (machine learning) to identify malignant tumours in histological images.
Career #3: Machine Learning Engineer (Analyst/Developer)
Machine learning engineers form a major branch of the data science discipline. They possess a rich knowledge of the various machine learning models that are available, and have developed an intuitive understanding of what models are best suited for a given task.
Professionals in this niche can work as either analysts or developers. Analysts use machine learning models, such as decision tree and k-nearest neighbours models, to categorise and find relationships within data. Developers create models that power services to carry out complex tasks, such as chatbots that use natural language processing models to answer user questions.
The developer and analyst paths require different types of work. Analysts don’t need to focus on developing a viable product, and so focus primarily on creating the most accurate and useful models possible. In contrast, developers need to create “minimum viable products” that must meet certain performance requirements (e.g., robustness), and so spend comparatively more time doing programming work.
Career #4: Data Engineer (Developer)
Modern organisations receive data from a variety of sources, such as point-of-sale terminals, online sales portals, marketing efforts, supply chain orders, etc. In many cases, these different sources will deliver data in a variety of different formats. Data engineers specialise in creating the infrastructure necessary to facilitate the movement of data from disparate sources into a common data store or analytics platform.
Data engineering is vital to the work of any organisation that ingests data from a large number of sources or in high volumes. A key advantage of data science over traditional analytics is the ability to analyse full datasets instead of working with subsamples. Two key types of infrastructure that data engineers work with are “data pipelines” and “data warehouses”, which move and store data respectively.
Data engineers also develop pipelines for moving data into analytics platforms, a type of operation called “ETL” (Extract, Transform, and Load). The ETL process transforms data into a format that is usable by a specific analytics platform without disrupting the existing data warehouse.
Due to the nature of their work, data engineers have sophisticated programming abilities, but put less emphasis on statistics skills than other data scientists.
Career #5: Data Warehouse Architect (Developer)
The data warehouse architect role is complementary to the data engineer role, and focuses on defining how an organisation’s data will be integrated, stored and accessed. Because differences in data structure affect how data can be accessed and used by analysts, data warehouse architects must ensure that complex datasets are stored in a manner that is suitable for their employer’s analytics needs.
To meet those needs, a data warehouse architect must understand what type of data storage (e.g., a relational, wide-column, graph or document-oriented database) to use for a given task. The ideal type of data storage to use in a warehouse depends on the particular analytics workflow that the warehouse is being used to serve—a warehouse that receives large amounts of data once-per-day (called “batch loading”) has different requirements than one which receives a constant stream of data.
Professionals in this field may also design non-warehouse data stores, such as knowledge graphs, which organise information in a manner that represents the relationships between each data point in the graph (database). This is useful for fields such as finance, as it enables analytics that show how a change in the fortunes of one company may impact the fortunes of the companies it does business with.
Career #6: Investigations & Data Journalism (Analyst)
The use of data science to perform investigations plays an increasingly important role in industries such as law enforcement, journalism, insurance, due diligence and risk assessment. Traditional investigations rely on subjective judgment to assess the significance of data, and must identify relevant connections within that data via manual inspection. Data science enables these analyses to occur at a larger scale, and allows significance to be efficiently assessed using statistics and other empirical measures.
The International Consortium of Investigative Journalists’ (ICIJ) investigation into the “Panama Papers leaks” is a prominent example of investigative data science. The ICIJ moved data from the 11.5 million leaked documents into a graph database, and performed analytics queries on the results to find relationships within the data. Through this process, they were able to expose how a number of individuals from jurisdictions around the world were hiding money offshore.
Data science offers an array of career opportunities
The above list of data science degree jobs is only a subset of the options that are available to qualified data scientists. The demand for data scientists has remained high in-part because data science skills are useful for a wide array of purposes, and data scientists continue to innovate new uses for their skills.
The University of New South Wales’ 100 per cent online Master of Data Science offers an ideal route forward for any individual interested in a data science career. As the only university in Australia that is globally ranked in Economics, Mathematics, Computer Science and Statistics, students in the program receive a top-tier education and will be among some of the most employable graduates in Australia.
UNSW data science students benefit from a curriculum that has been designed to teach the skills employers value most. Because the program’s online format provides the flexibility to structure your study around your existing life responsibilities, it’s suitable for those at any stage of their career.
The wide array of career opportunities available to data scientists make it ideal for anyone who wants to join an in-demand profession that offers its practitioners the flexibility necessary to chart their own course.
For more information about what UNSW’s data science programs have to offer, contact our enrolment team at 1300 974 990.