Dyna Software this week added a set of artificial intelligence (AI) agents to its portfolio that it has trained to configure application deployments on the ServiceNow software-as-a-service (SaaS) platform.
Launched at the Knowledge 2026 conference hosted by ServiceNow and scheduled to be available in the third quarter, the Platform Copilot developed by Dyna Software turns business requirements described in natural language into configurations.
Integrated with the application development tools provided by ServiceNow and large language models from OpenAI and Anthropic, the AI agents can create configurations using either text or images provided in, for example, a diagram.
Additionally, the AI agents will provide a preview of any proposed configurations before changes are applied along with audit history at the user, project and organization levels to capture every interaction and change produced.
Finally, there is also a user feedback mechanism that can be used to improve the quality and consistency of configurations.
Dyna Software CEO Ron Browning said the overall goal is to make it simpler for organizations to customize their ServiceNow application environments without having to rely on manual configurations that need to be created by an application developer. The goal is not to eliminate the need for those application developers but rather enable them to more adroitly respond to new requirements in a way that also serves to improve the quality of custom configurations, he added.
That capability makes it simpler for organizations to adjust configurations more confidently in ways that align more closely to the way they prefer to operate versus always requiring the business to operate the way ServiceNow has defined processes within its platform, noted Browning.
Most organizations to one degree or another wind up customizing SaaS applications. The challenge is that those customizations can be costly to develop and maintain. Dyna Software is now making a case for being able to not only reduce the cost of creating configurations but also more easily change them as workflows evolve in a way that also reduces potential risks to the business.
It’s not clear to what degree AI agents might enable more so-called citizen application developers to customize SaaS application environments. While many end users have been using any number of no-code/low-code tools to build applications, few of them adhere to best practices. As a result, many of those applications have security flaws, typically don’t scale and often provide suboptimal end user experiences. In theory, however, AI agents should make it simpler for end users who generally have a better understanding of business requirements to configure higher-quality applications.
AI agents should also eventually make it easier for organizations to build and deploy custom applications that span multiple SaaS application environments, with Dyna Software now eyeing adding support for Salesforce application environments, noted Browning.
Regardless of approach, the way applications are built and deployed has now been forever changed in the age of AI. The issue may not be so much whether to customize a SaaS application environment, but to what degree.
