AICoE Content Pipeline

This repo exists to effectively create, track, and manage the content created by the AICoE.

At AICoE, we have been actively involved in collaborating with teams within Red Hat to advance different products, services and operations using AI. In order to enable and/or educate teams and get engineers or stakeholders more involved in the projects, and in order to get them to be able to effectively use the solutions, we need to streamline the process in which we make the engagements and have a way to measure the impact of the project.

Objective

• Enable engineers on the teams we are engaging with to try out the solutions.
• Receive feedback whether the solution is effective or not.
• Measure impact created by the collateral.
• Enable external teams to use AICoE tools in their development cycle

Traits of artifacts

• consumable, reproducible structure.
• Leverage Reproducible DS workflow standards.
• findings and results in a understandable and discoverable format
• Notebooks and projects are easy to understand for the inexperienced
  • Sample data is included
  • Time boxed: each chapter should not exceed 15-20 minutes
• make project onboarding easy by following the standards
• democratize data access to empower all stakeholders to experiment with data.
• Create upstream versions of the same artifacts with sample data or public data

Impact

• Advertise possibilities and artifacts at various outlets
  • blog, community of practice, mailing lists, tech talks
• Measure impact created by the project
  • track use of project
  • track PRs/contribution/feedback surveys/issues created
• Measure impact created by artifacts
• gather feedback from stakeholders and determine how we can involve more people.