GitLab this week revealed that, in the coming months, it will add an enterprise edition of its artificial intelligence (AI) add-on for its namesake continuous integration/continuous delivery (CI/CD) platform.
Duo Enterprise combines the existing ability to make code suggestions and maintain privacy controls using large language models (LLMs) with additional capabilities that enable DevOps teams to proactively detect and fix security vulnerabilities, summarize issue discussions and merge requests, eliminate bottlenecks and enhance team collaboration.
A dashboard also provides a value stream forecasting capability to provide insights into its usage of AI features and their impact on the metrics being tracked by a DevOps team.
Additionally, GitLab Duo Enterprise provides the option for self-hosted model deployments to support organizations that cannot connect their secure, air-gapped environments to internet-enabled services.
Paul Nashawaty, practice lead for application development and modernization at The Futurum Group, said capabilities such as vulnerability explanation and automated remediation will further the usage of AI across the software development lifecycle (SDLC). At present, only 18% of organizations use AI in production applications, he noted.
Separately, GitLab this week also made generally available a CI/CD catalog to provide a centralized portal through which DevOps teams can discover, reuse, and contribute pre-built CI/CD components. In addition to the public catalog, organizations can create a private catalog to distribute customized pipelines that automate specific workflows.
In addition, GitLab is readying a GitLab 17 update that will include observability tools, project planning capabilities, a secrets manager that runs natively and integrations with static application security testing (SAST) tools.
That update will also add a registry to enable data scientists to develop AI models on the same platform where engineers build, test, secure and deploy code.
Finally, GitLab is making an edition of its CI/CD platform for the Google Cloud Platform that runs in a private cloud to address compliance requirements.
Apply AI to Manage DevOps Workflows
As more developers rely on AI tools to write code the overall size of the codebase that DevOps teams will be required to manage is only going to increase. As that occurs, DevOps teams will need to apply AI to help manage the DevOps workflows through which that code eventually winds up being deployed in production environments. Most organizations won’t be able to hire a small army of software engineers so the only way to rise to the challenge will be to rely more on AI. Less clear is to what degree that requirement will drive organizations to replace their existing CI/CD platforms in favor of more modern alternatives that have AI capabilities.
In the meantime, however, DevOps teams should be preparing now for an AI era where many of the routine tasks that today conspire to slow down the pace at which applications are built and deployed are eliminated. The challenge then becomes determining how DevOps teams augmented by AI tools will revamp their existing workflows to enable more applications to be simultaneously deployed and updated at a level of scale that a few years ago would have seemed unattainable.