AI-assisted coding — a practice that uses artificial intelligence to help generate code — benefits software developers first and foremost.
But the potential benefits of AI-assisted coding aren't limited to software development. They could also extend into other IT domains, including IT operations. Indeed, going forward, the most efficient ITOps teams will likely be relying increasingly on AI-powered coding tools to help speed their work and reduce the risk of making mistakes.
To prove the point, here's a look at four areas where ITOps engineers can benefit from AI-assisted code generation tools and platforms, along with an estimation of how close we are to a world where AI-powered coding tools become a routine part of IT operations workflows.
1. Creating Infrastructure-as-Code Templates
Infrastructure-as-code (IaC) tools are a great solution for ITOps engineers who have to configure large numbers of servers or cloud hosting environments. By writing code to define how the resources should be configured, then applying the configurations automatically, ITOps engineers can use IaC to avoid having to configure each resource by hand — a time-consuming and error-prone process.
However, the caveat with IaC is that someone has to write the IaC templates that determine how resources will be configured. Traditionally, that's a task that ITOps engineers have had to perform by hand.
But in the future, configuring IaC files may no longer require manual effort. AI-powered coding tools could write the files automatically, based on natural-language input from ITOps engineers about how infrastructure should be configured. Rather than writing, say, a Terraform file by hand, you could simply say, "Write a Terraform file to set up a t2.micro instance," and an AI tool would write the necessary configuration code for you.
2. Writing Policy-as-Code Rules
ITOps teams could adopt a similar strategy for writing policy-as-code files. Policy-as-code is an approach to setting up IT resources of any type — not just infrastructure — using code-based configuration files.
As with IaC files, someone has traditionally had to write policy-as-code rules or policies by hand. But AI-assisted coding could allow ITOps teams to generate policy-as-code rules automatically.
This would be important because it might encourage further adoption of policy-as-code approaches to IT operations management. Today, the use of code-based configuration engines remains primarily limited to IaC. But if it were possible to generate all types of policy configurations automatically, it would become easier to configure any type of IT resource using this approach.
3. Alert Management
Alert management is another area where ITOps teams can benefit from AI-assisted coding. To manage alerts, engineers have traditionally had to write rules that determine when a monitoring tool should generate an alert, as well as how interrelated alerts should be grouped together.
AI-powered code generation tools could automate this process. Engineers could describe how alerts should be managed, and the tools could write code to manage the alerts accordingly.
4. Configuring Observability Tools
Likewise, observability tools could conceivably be configured using code generated by AI-powered tools. ITOps engineers could describe the metrics they want to collect and the data sources they want to monitor, then leave it to AI to write configuration files for their observability platforms based on those descriptions.
We Have the Technology
Technologically speaking, the tools necessary to perform the types of work described above already exist. For example, here's the output I received when I asked OpenAI to "generate a Terraform file to create a t2.micro instance":
provider "aws" {
region = "us-east-1"
access_key = "ACCESS_KEY_HERE"
secret_key = "SECRET_KEY_HERE"
}
resource "aws_instance" "test" {
ami = "ami-0ff8a91d"
instance_type = "t2.micro"
}
Admittedly, that was a pretty simple query, but it shows that – at least for basic use cases – the AI technology available today is already quite good at writing the code necessary to complete ITOps tasks.
On the other hand, there are limitations on the extent to which today's AI-assisted coding technologies can address real-world ITOps needs. One problem is that, for the most part, the AI-assisted coding tools available today, like GitHub Copilot and Visual Studio IntelliCode, are geared toward software development more than ITOps, and they are generally better at writing software code than at writing configuration code.
A second challenge is that a lot of the ITOps platforms that could benefit most from AI-assisted coding don't yet support the integrations necessary to enable AI-powered configurations. In other words, we aren't yet seeing integrations between, for instance, observability platforms on the one hand and AI-assisted coding tools on the other. Vendors that target the ITOps space will need to embrace AI-assisted code generation in order for this to change.
Still, it seems pretty clear that AI-assisted coding already offers tangible benefits to IT operations teams, in addition to software development teams. The challenge going forward is for ITOps to figure out how to leverage AI-assisted coding in an efficient way.
About the author
Christopher Tozzi is a technology analyst with subject matter expertise in cloud computing, application development, open source software, virtualization, containers and more. He also lectures at a major university in the Albany, New York, area. His book, “For Fun and Profit: A History of the Free and Open Source Software Revolution,” was published by MIT Press.
