How businesses can opt for a more autonomously engineered stack

Cloud computing concept represented by a server room, with a cloud representation hologram concept.
(Image credit: Shutterstock/Virgiliu Obada)

Automation technologies have become a vital part of business operations, and they are continuing to accelerate and permeate all levels of business in a variety of industries. Companies across the healthcare, financial services and transport industries, to name a few, are starting to quickly identify an increase in software, surface and process test automations, as well as increased development of no code apps.

However, navigating the automation market can be daunting due to the plethora of automation processes and developments available. Yet, with the emergence of new processes and AI tools, businesses have the opportunity to experiment with automation in software, surface operations, and app development.

Kavitha Chennupati

Senior Director of Product Management, SS&C Blue Prism.

Software automation for software

Imagine automation suites that can automate the entire process development lifecycle for a business. These technologies can auto-discover processes, tasks, and user interfaces, generate test data, create test cases, and deploy tested processes, leading to automation co-development or developers-in-the-loop.

The auto-discovery of automated processes has developed relatively quickly in recent years. Previously, process discovery was relatively time-consuming, resulting in companies conducting workshops with several business users to identify processes, business rules, and an efficient flow of operations. Process orchestration was thereby defined, and business rules were plugged into the process. Once process orchestration was defined and incorporated within the wider value chain, businesses set about ensuring that this was followed by thorough documentation for all businesses.

Nowadays, modern process intelligence tools have the capability to automatically suggest how to automate certain processes and rules by analyzing and understanding the flow of those processes. The process model can then be the basis for shorter-duration workshops to ratify the flow and develop the ground-up processes or reengineering leading to faster automation. Process documentation can also be auto-generated using natural language tools aka GenAI/LLMs.

As the technology continues to advance, auto-reengineered processes can be deployed and will be executed throughout a set period of time. During , the execution there could be collaboration with human worker for review of data and results before completing the process. This further reinforces the idea that automation and human employees can work in tandem, complementing each other's work to enhance their business's automation capabilities.

Surface automation

Previously, building automation based on RPA software required automation developers to manually identify form elements and script them into the automation processes. As the AI race gains momentum, recent advancements in artificial intelligence (AI) have enabled AI-based surface automation to intelligently capture screen elements, automatically generating robotic processes. This leads to faster automation processes. Automation stacks in the future will soon be able to generate robotic processes through Computer Vision, auto-define the orchestration of several workflows, and self-govern task configurations. This will allow developers to employ auto-discover processes, therefore freeing their time to focus on more complex orchestrations and configuration processes.

No-code app development and form snippets

Before the time of fast app development, creating applications meant several months spent gathering requirements from businesses, documenting approvals and onboarding developers with specific business requirements. Developers had to then schedule resources, and write, and test code before commencing deployment. The delivery of these developments could then be further extended when delays occur in resource allocation, unclear requirements, or a shortage of skills.

Nowadays, business users can generate apps by themselves using no-code app development tools and deploy them in a matter of hours or days. However, business users still require further assistance for app governance and management.

So, with no code app development here to stay, business users will begin to see natural language-based, voice-based, computer vision-based or text-based form generation. For instance, the business user says, "generate a form for claims processing that can take input as claim number and claim form" and then the form design flashes with appropriate datatypes and variables associations.

As business users dual up as citizen developers, automation developers can leverage progress by gaining domain knowledge and building a base library of assets that can be reused, including data models to leverage in forms. Developers also have the capability to enable automated testing and deployments, as well as establish governance and regulation over automation and AI platforms within a business.

Process test automation

Testing operation processes has previously been a manual activity, repeating through several hundreds and thousands of paths, supplying relevant test data, and iterating through test cases manually for each process. This long, time-consuming activity can often take hours of work, whereas now, test cases can be auto-generated. Developers can capture test data from the first manual iteration, and this can then be repeated during upgrades. Therefore, automation developers will no longer require specialized test automation skills and can verify the coverage and test reports with their business skills.

In the future synthetic test data can be generated to simulate and test the processes and verify the configuration accuracies. This will allow the automation developer to come into the loop to verify the test report and pay attention to only failed test cases that need fixes. Developers will soon find their roles transforming into those of dev-test-deploy automation developers, emphasizing complex problem-solving tasks. This includes investigating the reasons behind failed test cases and determining necessary process adjustments or adaptations to rectify them.

Automation of Dev-Test-Ops

Automation developers have previously been plagued with a large stack of manual tasks. Developers had to manually migrate the processes from one environment to another, sometimes even re-doing some configurations. The scale of moving these processes were often very time-consuming and involved several hours of work, resulting in less time for more complex tasks.

Nowadays, developers have one-click auto-test and deployment leading to quicker release of features to the clients, also resulting in faster ROI for companies.

While automating dev-test-ops can speed up manual processes, the automation industry will soon see an increase in auto-scaling and sizing the infra with dynamic billing based on the process needs. Process telemetry and process or mining data can result in self-learning processes, allowing automated platforms to learn as they develop. This will allow developers to gain skills like collaboration with stakeholders and domain language and be prepared to be a multi-faceted generalized automation developer.

Furthermore, the notion that these processes can continually adapt and growth as business demands change and fluctuate demonstrates the flexibility of how automation can propel businesses into the future.

We've listed the best DevOps tools.

This article was produced as part of TechRadarPro's Expert Insights channel where we feature the best and brightest minds in the technology industry today. The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here:

Kavitha Chennupati, Senior Director of Product Management, SS&C Blue Prism.