“Not only are there no silver bullets…”
The Mythical Man-Month: Essays on Software Engineering
This chapter explores scenarios that organizations may encounter on their journey toward leveraging artificial intelligence (AI) for predicting, automation, and optimization.
The scenarios are viewed across the contexts of various xOps approaches as a means toward continual operational improvement.
xOps is a shorthand way to indicate various types of operational disciplines that are helpful for managing robust analytical platforms for AI and other forms in an agile and sustainable manner. These operational disciplines are as follows:
- DevOps and MLOps—development and IT operations
- DataOps—data operations
- AIOps—AI for IT operations
DevOps and MLOps involve a combination of software development practices with IT operations to help shorten the amount of time to develop and release software and ML/AI models into production. DataOps focuses on the processes that improve the speed and accuracy of analytics, including data access, quality control, automation, integration, and model deployment for AI. AIOps combines DevOps with machine learning and AI to help drive faster root-cause analysis and accelerate the mean time to repair. AIOps uses longitudinal operational data to help identify signals that can indicate a negative situation.
Each xOp is going to embrace its own nuanced methodology. The Ladder to AI ...