Hyperautomation meets DevOps

Historically, as humans we try to find ways to do things faster and in a more efficient way, saving ourselves some energy for doing better things. Invention of wheel by our forefathers and early man’s tools inventions were the earliest form of automation. Since then the quest to be better and faster is ongoing.

Similarly, in software development, automation has always been the key driving factor. There have always been efforts to automate mundane and repetitive tasks across SDLC. It has both human and business perspective to it. Smart IT engineers do not find it intriguing to spend hours doing a manual code review or test an application manually, rather they would find interesting ways to automate these monotonous and routine tasks. Even from business perspective, it improves the efficiency and hence speed to market.

As technologies advanced and the pace of Devops adoption picked up, it started becoming evident that it’s not possible for one technology to automate a complete process that a single human can do. There are multiple disparate system interfaces and human interventions that are needed. Thus, the idea to combine different tools that will enable automating tasks which couldn’t otherwise be automated earlier came into play. Hyper automation became mainstream after it got it’s mention in Gartner Top 10 Strategic Technology Trends for 2020.

Hyperautomation is the application of advanced technologies like RPA, Artificial Intelligence, machine learning and Process Mining to augment workers and automate processes in ways that are significantly more impactful that traditional automation capabilities. (definition from Wikipedia)

Hyperautomation covers the vast technologies of  –

1.      RPA

2.      Process mining

3.      NLP

4.      AI

Hyperautomation has its application in all industries – Anti Money laundering, Banking, customer onboarding, order management, and is also started to see its adoption. Here is an example of how hyper automation can be applied in a Devops delivery model to augment and accelerate software delivery -

Dev – AI to review user stories and design, developer reusing existing snippets of code from a central store, automating unit tests, intelligence built in IDE to guide developers as they write their code, building end to end pipeline through Jenkins and pipeline templatization.

Test – System and acceptance tests using Selenium or similar tools, intelligent API automation, using RPA tools to automate processes which could not be automated by traditional automation tool suites, automate discovery and generation of test data, service virtualization and use of AI/ML techniques to identify tests to be executed

Ops – Infra as a code, release automation and automated environment management by use of containers. Real time continuous intelligent feedback is the key across this journey. This allows team to stay connected and help them move to a proactive rather than reactive mode of issue resolution.

Hyperautomation should help improve agility and efficiency to a power of 10x over conventional automation practices. Success of hyperautomation also depends on ability of identify right set of tools that compliment each other. One of the key considerations to enable in the hyperautomation framework is to keep it technology agnostic and provide for an ability to plug in different tools on need basis.

Hyperautomation not only gives a competitive edge but also frees up human capacity which can be used for making high order decisions and applying common sense.