Introducing Spacelift Intelligence: AI That Lives Inside Your Infrastructure Workflow

There is a gap forming inside most engineering organizations, and it is not a tooling gap. It is a velocity gap.

Development has shifted to AI-augmented workflows. Engineers use AI to write code, reason about systems, and move faster. Infrastructure, meanwhile, remains artifact-first and pipeline-bound. Every change still flows through a commit, a plan, a review, a merge. That process exists for good reasons. But when development velocity increases dramatically, relying exclusively on that path creates friction, and the friction is growing.

This is not a criticism of IaC or GitOps. Those practices are foundational. The engineers who built them made the right call for the era they were operating in, when infrastructure changes were deliberate, infrequent, and high-stakes. The problem is not the tools. It is the mismatch between operating models. Development has moved. Infrastructure workflow, as a category, largely has not.

Today, we are changing that with Spacelift Intelligence.

Spacelift Intelligence is not an AI feature bolted onto the platform. It is a new product layer built around a simple premise: AI should participate directly in infrastructure workflows, not just produce artifacts that feed into them.

It ships with two components that work together: an AI Infrastructure Assistant and a natural language deployment. model.

Infra Assistant is a conversational AI interface embedded directly in Spacelift. It understands your stacks, your state, your runs, and your configuration. You can ask it questions about your infrastructure, get answers grounded in what Spacelift actually knows about your environment, and take action without leaving the platform. It is not a simple chatbot that generates Terraform and hands it back to you. It is a workflow participant that has context.

• “Please find any public s3 buckets across all my stacks”
• “Do I have any CVE vulnerabilities in my environment?”
• “What would you recommend I do to optimize my production deployments?”

Spacelift Intent goes further. Intent lets you provision and manage infrastructure through natural language, with Spacelift handling the orchestration end-to-end. You describe what you want. Spacelift Intent figures out what that means in infrastructure terms, plans the change, and executes it. Your policies still apply. Your approval flows still run. All with no code. Your state is still tracked. The operating model changes. The control surface does not.

• “I need an S3 bucket to store some temporary files. Adhere to any company policies.”
• “Please create and test a policy to not allow public ec2 instances.”
• “Please build 3 ec2 stances in us-west-1 so I can test a prototype.”

Together, these two components form a layer that connects natural language to real infrastructure action and insights, inside Spacelift, against real environments.

Most vendors exploring AI for infrastructure have converged on the same pattern: help engineers write Terraform faster. That is useful, but it’s also incremental. Generating code faster does not change the fact that the code still has to travel through every stage of an artifact-driven pipeline before anything happens.

The bottleneck in infrastructure workflows is rarely the writing of Terraform. It is the process that surrounds it. The cycle from intent to deployed change, through commit, CI, plan, review, merge, and apply, was designed for a pace of work that many organizations have already left behind.

Spacelift Intelligence is designed to address that bottleneck directly. The goal is not to accelerate code generation. It is to give AI a seat at the table in the workflow itself, so that the distance between what an engineer wants and what gets provisioned is measured in interactions, not pipeline stages.

IaC is not going away. GitOps is not going away. Your existing Spacelift stacks, Terraform code, VCS integrations, policies, and governance model stay exactly as they are.

Spacelift Intelligence extends what is already there. Some changes warrant the full pipeline and rigor. Some do not.

Having both options, governed by the same control plane, is where Spacelift Intelligence shines. Engineers and platform teams who want to continue working through Git-driven pipelines can do that without any changes. Intelligence just adds a new path: one where AI can understand, design, deploy, and govern infrastructure.

Spacelift Intelligence is built for platform engineers, DevOps leads, and infrastructure architects. The people responsible for making infrastructure workflows work at scale, for multiple teams, across complex environments.

The Infrastructure Assistant gives those teams faster access to what Spacelift knows: the state of stacks, history of runs, and shape of your infrastructure. Spacelift Intent gives them, and the teams they support, a way to act on that knowledge without forcing every change through the same artifact-driven process.

The underlying platform, workers, policies, spaces, and governance model do not change. What changes is how you can engage with it.

The framing of AI for infrastructure has, until now, been primarily generative. Write the code faster. Scaffold the module. Suggest the fix. That framing treats AI as a pre-pipeline accelerant.

Spacelift Intelligence treats AI as a workflow participant. It operates inside the platform, with access to real infrastructure context, and it executes through the same machinery that governs every other change. That is a different kind of integration, and it reflects a different view of where AI belongs in the infrastructure stack.

Not upstream of the workflow. Inside it.

Spacelift Intelligence is available now! We’ve put together a quick demo video where you can see it in action below:

If you’d like to learn more, you can sign up for our upcoming webinar, where we will discuss Spacelift Intelligence in depth here!