An AI-native infrastructure agent that makes configs write themselves

Cloud infrastructure is still too cumbersome. Most teams rely on brittle CI/CD workflows, manual configuration, and DevOps bottlenecks just to make simple changes. Infrastructure-as-code (IaC) has moved things forward, but it hasn’t kept pace with how modern software is built today or software that is now built with AI. 

Innovation in infra is slow because it’s complex, it’s usually built on legacy systems that are hard to modernize, and it must be stable and secure. It cannot go down without wreaking havoc on businesses. This leads to risk aversion in the space as well. Digger is changing that.

The team at Digger has been building for infra for a long time and have already had a hand in shaping its evolution. Digger began as an open source change management system for teams using tools like Terraform. Today it’s had half a million downloads and powers deployments for over 300 companies, from major banks to casinos to freight companies, helping them manage infrastructure with automation and precision.

Building for infra with AI

Now Digger is going a step further. They’ve launched Infrabase, an AI-powered DevOps agent that lives in the pull request. It automates and simplifies the lifecycle of an infrastructure change, but with the same rigor, control, and visibility that is critical for enterprises.

Now that LLMs are capable of generating reliable config, which most developers don’t want to write, DevOps can prompt instead of writing the config code. Digger then provides the control layer organizations need that manages access, enforces policy, flags risk, and keeps a clear audit trail, all while submitting updates automatically, before anything hits the cloud. Security, compliance and access control are all checked up front.

Even more exciting is the bigger future Digger is building toward: infrastructure that writes and reviews itself, triggered by a prompt. AI systems can generate application code and Digger can deploy it safely and at scale, giving infra engineers superpowers.

Why this team

The Digger founders have deep experience with infrastructure and were working on AI-generated config before LLMs were mainstream. 

Igor Zalutski and Mohamed Habib were inspired by internal self-service infra tooling at Palantir, Amazon and Fitbit, and Utpal Nadiger helped launch OpenTofu, a LinuxFoundation-backed fork of Terraform. The founders have built a cloud management UI for startups on top of Terraform, and originally built Digger, the popular open source runner for Terraform, as a side project. 

We were impressed with the depth of their experience, the traction of the open source product, and how they are using AI to shape the next phase of how infrastructure is built and shared, and we can’t think of a better team to do it.

We’re proud to back Digger as they take infrastructure into the future, building the infrastructure layer AI-generated software needs. 

They’re growing their team. If you’re interested in joining them, check out their job openings here.