While there is strong focus on the role of the data scientist specifically, successful enterprise machine learning is a joint effort, with team members drawn from IT, analytics and the relevant business function, and led by a strong project manager, all contributing towards a single solution/product. People from the business side understand the nuances of the problem at hand, data engineers know where and how to get and prepare the required data and make it available for the machine engineer, who will do the modeling, etc.
Below are the key roles; actual titles vary from organization to organization. The size of the team and its actual constitution depend on the scope of the project, the organizational structure, the capabilities of machine learning platform employed, and the skills of the participants; a small project can be accomplished with few participants who take up multiple roles, while a large project may require multiple participants to fill the same role.
Project Sponsor: Initiates and evangelizes the project within the organization. Defines the scope of the business problem to be solved and the desired results, and evaluates the extent to which project results meet the business objective. Ensures that the project is aligned with strategic goals and that project results will be implemented.
Project Manager: Ensures that the project stays on schedule, is completed within schedule, and key milestones and objectives are delivered at the expected quality.
Business User: Businessperson who cares about problem, can quantify the business value of potential solutions and will use the solution if is implemented.
Domain Expert: Usually filled by multiple participants such as business analysts, business intelligence analysts, operational managers, and line-of-business subject matter experts. Consults and advises the project team on the project context, the value of the results, and how the results will be applied to the business. Involvement increases the probability of project success and application of results to the business.
Database Administrator: Sets up and configures the databases to be used, providing appropriate access to key databases and tables to team members.
Data Scientist: Builds machine learning models for solving the business problem by applying appropriate techniques. Ensures that machine learning objectives are met.
Data Engineer: Provides the infrastructure and expertise for data collection, transformation/manipulation and storage.
Data Developer/Software Engineer: Provides the code and automation that enables end user access to deployed machine learning models.
Machine Learning Engineer: Builds tools for model deployment and deploys models in the cloud, on-premises or at the edge as appropriate.
Citizen Data Scientist: Uses powerful software and domain expertise to create routine machine learning models without having received formal training in advanced mathematics, statistics or data science, freeing data scientists who have additional skills to handle complex projects.