The DS3 Workbench forms the umbrella for inward-facing projects that build infrastructure for Data Science & Software Services (DS3) operations. The main goal is to support intake and management of faculty research projects that are supported by DS3. In particular, the workbench will provide to following functionality:
- an interface that allows researchers to submit project proposals and resource requests,
- send notifications to researchers about potential grant opportunities for their projects,
- an interface that allows students to submit their CV and express their interest to work for DS3 in the future,
- assist DS3 administrators with resource and project planning, and with the hiring process.
An illustration of the overall architecture is shown below.
Overview of the DS3 Workbench architecture.
Researchers can enter their project proposals and resource requests via a customer user interface. Among other information they need to specify the kind of work they expect from DS3 as part of their project (e.g., required skills and nice to have skills). If a researcher proposes a project for which they currently do not have grant funding the workbench can notify them about suitable grant opportunities.
The matchmaking tool assists with various tasks: identify (a set of) employees to work on a submitted project, highlight shortage of skills to satisfy project requirements, identify potential candidates for hiring.
The grant recommendation engine uses GrantForward (a service that the University can subscribe to) for continuous queries (e.g., once a day) to identify grant opportunities for projects that do not yet have funding.
Follow the instructions on the Development Setup page to setup an instance of the workbench Web API. The user interface is currently distributed across two repositories:
- External UI: External facing user interface for submitting project proposals and registering student workers.
- Admin UI: Internal facing user interface that provides functionality for DS3 administrators to view match making results and manage worker allocations.