Spacelift, this week, added generative artificial intelligence (AI) capabilities to its platform for automating the management of IT infrastructure provisioned using code.
Company CEO Pawel Hytry said Saturnhead AI adds a set of AI capabilities that can, for example, be used to identify issues and suggest potential remediations to make it easier for DevOps teams to troubleshoot complex IT infrastructure environments.
Compatible with multiple large language models (LLMs) that DevOps teams can specify, Saturnhead AI is able to analyze infrastructure logs in real time to explain in natural language what has happened and why it occurred. That capability eliminates the need to otherwise manually sift through those infrastructure logs in IT environments where there might be thousands of failed runs per week, noted Hytry.
While many DevOps teams routinely make use of infrastructure-as-code (IaC) tools, managing IT infrastructure remains challenging. A recent Spacelift survey found that only 14% of respondents have fully implemented infrastructure automation best practices. Generative AI tools have the potential to enable more organizations to more consistently apply best DevOps practices in an era where IT infrastructure has become too complex for humans to manage, noted Hytry.
Less clear is the degree to which AI might enable junior members of a DevOps team to effectively manage IT infrastructure at scale versus allowing senior members of a DevOps team to manage IT infrastructure at much higher levels of scale. The one certain thing is that it will become easier for DevOps teams to share tribal knowledge about how infrastructure is configured, said Hytry.
Each organization will need to determine how best to apply AI to managing IT infrastructure, but there will always be a need for humans to supervise AI models, said Hytry.
Spacelift, while still investigating the potential role AI agents might play in automating the management of IT infrastructure, is pursuing a more conservative approach given the number of errors that might be injected into a DevOps workflow, said Hytry. Most IT operations teams are going to be reluctant to rely on any form of automation that might inadvertently make critical IT infrastructure suddenly unavailable to applications, he added.
A recent Futurum Research survey finds 41% of respondents expect generative AI tools and platforms will be used to generate, review and test code, but the degree to which AI will be relied on to manage IT infrastructure is a little less clear. The one certain thing is that IT infrastructure is only going to become more challenging than ever to manage. Unfortunately, many of the processes being employed by application developers are far more manual than they should be, which creates more opportunities for misconfigurations. As a result, application developers are spending time on a set of tasks that, in addition to possibly introducing cybersecurity risks, also take time away from writing code.
Hopefully, advances in artificial intelligence (AI) will soon make it easier to discover and remediate these issues faster. In the meantime, however, DevOps teams would be well advised to invest more in automation to ensure infrastructure is not only being provisioned faster but also more reliably.
