New: LLM Rankings are live!
See how the top LLMs rank on real infrastructure tasks
How Claude Fable 5.1 did on iac-bench 2026-10: an overall composite of 86.9, scores per fixture, strengths, weaknesses and the patterns in its OpenTofu modules.
How Claude Fable 5.1 did on iac-bench 2026-10: an overall composite of 86.9, scores per fixture, strengths, weaknesses and the patterns in its OpenTofu modules.
How Claude Haiku 5.5 did on iac-bench 2026-10: an overall composite of 68.0, scores per fixture, strengths, weaknesses and the patterns in its OpenTofu modules.
How Claude Opus 5.5 did on iac-bench 2026-10: an overall composite of 88.9, scores per fixture, strengths, weaknesses and the patterns in its OpenTofu modules.
How Claude Sonnet 5.5 did on iac-bench 2026-10: an overall composite of 86.1, scores per fixture, strengths, weaknesses and the patterns in its OpenTofu modules.
How DeepSeek V4 Pro did on iac-bench 2026-10: an overall composite of 69.2, scores per fixture, strengths, weaknesses and the patterns in its OpenTofu modules.
How Gemini 3.1 Pro did on iac-bench 2026-10: an overall composite of 73.0, scores per fixture, strengths, weaknesses and the patterns in its OpenTofu modules.
How Gemini 3.6 Flash did on iac-bench 2026-10: an overall composite of 54.3, scores per fixture, strengths, weaknesses and the patterns in its OpenTofu modules.
How GPT-6 Astra did on iac-bench 2026-10: an overall composite of 86.7, scores per fixture, strengths, weaknesses and the patterns in its OpenTofu modules.
How GPT-6 Luna did on iac-bench 2026-10: an overall composite of 81.5, scores per fixture, strengths, weaknesses and the patterns in its OpenTofu modules.
How Kimi K3 did on iac-bench 2026-10: an overall composite of 82.4, scores per fixture, strengths, weaknesses and the patterns in its OpenTofu modules.
How the iac-bench 2026-10 results were produced: the briefs, the evaluation setup, mechanical scoring against real tofu plans, the calibrated LLM judge, statistics and limitations.
We asked ten language models to write reusable OpenTofu modules for AWS, GCP and Azure, then scored 90 modules against real plans and a calibrated judge. Here is how they did and the patterns behind the scores.
Reference for every criterion behind an iac-bench score: each rubric item, expertise item and judged doctrine section, with weights and how many attempts passed.
The Platform Engineering Podcast moves to massdriver.cloud, applications that run on runtimes your platform team owns, OIDC cloud credentials for deployments, custom default tags, proposal review in the config form, cloud resource import from Claude Code and opencode, search and sort for members and groups, service accounts you can inspect and delete, projects copied from an environment, and config forms that collapse long lists and check your input before you deploy.
Seats follow your identity provider, custom attributes are safe to change, config can be loaded and promoted from any instance, and external resources show on the environment graph.
Agentic SDLC and software factory guides cover planning through code review. Here is what agents need to release and operate infrastructure safely.
How release channels, versioned resource types, and version-aware links roll a module upgrade across every project and environment from one push.
Resource types become versioned OCI artifacts, connections carry version ranges, and deployment approvals get separation of duty.
Half your company is already routing around IT. The citizen developer has been around since Lotus 1-2-3, and guardrails are the only control ops has left.
Identity-based permissions break self-service. We tolerated it for twenty years because the breakage was slow. Agents remove that buffer entirely.
AI agents need real infrastructure to test against. Not plans. Not mocks. We are building disposable, isolated environments where agents do the labor and humans make the call.
AI coding tools work shockingly well. That's what makes the structural risks so hard to dismiss. A reflection on productivity, dependency, and the quiet consolidation of the intelligence layer.
Version 1.3.0 of the Massdriver Platform featuring SCIM 2.0, Integrations Console, and Claude Code Plugin.
Version 1.2.0 of the Massdriver Platform.
Infra changes ship faster when Day-2 stops being a guessing game — this release makes that real.
Struggling to scale Infrastructure as Code? This post explains why CI/CD pipelines and Git workflows break down for modern cloud infrastructure, how brittle IaC practices slow teams, and why a new foundation—pipeline-less delivery with built-in policy, compliance, and orchestration—can help DevOps scale and empower developers safely.
It’s a Symptom of Platform Design, Not Developer Behavior.
AWS forces users to handle the management of kube-proxy in EKS clusters. If you're experiencing strange network behavior in a long-lived EKS cluster, this could be the problem. Let's go over some solutions.
Most Infrastructure as Code implementations expose too much complexity, forcing developers to make infrastructure decisions they shouldn’t have to. This post explores why generic IaC modules fail, how to design abstractions that encode organizational standards while intentionally omitting irrelevant configurations, and how to shift ops from gatekeeping to enablement.
GitOps isn’t the right fit for modern Ops, DevOps, or platform engineering. Learn why static workflows fail to address dynamic cloud complexity and explore better tools for managing real-time infrastructure.
Operations engineering is stuck in the past. This post breaks down why ops hasn’t had the same revolution as software development, explores the limits of DevOps, and argues for scaling expertise through reusable tools and clean abstractions. It’s time for ops to adopt lessons from software engineering and move the industry forward.
Your Guardrails are Bullshit
Learn how to implement convention-over-configuration for Infrastructure as Code (IaC) using OpenTofu. Discover practical techniques to build developer-friendly IaC modules that streamline compliance, enforce business standards, and reduce operational friction.
Welcome to Week 10 of our OpenTofu Foundations series! This week, we’ll focus on essential tools that ensure your OpenTofu configurations are reliable, secure, and maintainable. By the end of this session, you’ll have enabled key tools, understood their value, and integrated them into your workflow.
In this webinar, we'll look at advanced strategies for using and manipulating the statefile. If you work with OpenTofu (or Terraform) long enough, you **will** have to repair, modify, or recover state. Guaranteed.
Welcome to Week 8 of our OpenTofu Foundations series! In this session, we'll delve into how OpenTofu manages the state of your infrastructure. Understanding state is crucial for effectively using OpenTofu, especially when working with teams and multiple environments. We’ll explore state files, secure storage options, and state locking strategies to ensure your infrastructure remains consistent and secure.
In this session we will learn how to manage multiple deployments with the same OpenTofu configuration. We will learn about workspaces which are OpenTofu's built in mechanism for managing multiple state backends. Next we will cover some tricks to make managing workspaces less error prone for users. Lastly we will look at Checkov and how to implement security and policy scanning to ensure your deployments meet your compliance needs.
Learn how to implement Convention over Configuration principles in your Infrastructure as Code using OpenTofu, making your IaC more maintainable and developer-friendly.
OpenTofu is an infrastructure-as-code tool that lets you define both cloud and on-prem resources in human-readable configuration files that you can version, reuse, and share. This post will help you get started using OpenTofu to manage your Azure resources.
Learn how to incorporate testing and validation into your OpenTofu workflows. This session will focus on tools and practices for testing your infrastructure code, including syntax validation, resource policy checks, and simulating infrastructure changes before applying them.
Integrate OpenTofu into CI/CD pipelines to automate infrastructure deployments. Set up a pipeline to test, validate, and apply configurations reliably.
Welcome to week four of the OpenTofu Foundations workshop! This week, we'll dive into some of the advanced strategies for managing OpenTofu resources, including data sources, loops, conditionals, dynamic blocks, and lifecycle rules.
Welcome to week three of the OpenTofu workshop This week, we'll dive into functions and control structures. 90% of what we do as programmers is validate unknown data being passed to our programs, and manipulating imperfect data so we can connect software modules together. Functions help us provide a better more intuitive user experience and keep our code clean.
As platform engineers, DevOps professionals, and operations experts, our primary role is to empower developers—to make them more efficient, to help them ship faster and more confidently.
.png&w=3840&q=75)
Welcome to week two of the OpenTofu workshop! This week, we'll dive into the concept of modular infrastructure using OpenTofu. Modules are a powerful way to organize your code, promote reusability, and enforce best practices in Infrastructure as Code (IaC). By the end of this session, you'll be able to create your own modules, use existing ones, and apply best practices for modular infrastructure design.
Welcome to week one of the OpenTofu workshop! In this module, you'll learn how to create an EC2 instance running Wordpress, set up a MariaDB database, and configure basic infrastructure in AWS. We’ll go through creating instances, security groups, and databases while discussing key Terraform/OpenTofu concepts.
Are you ready to dive into the world of Infrastructure as Code (IaC) with OpenTofu? We're excited to introduce this 10-week series, designed to take you from a beginner to a confident practitioner of OpenTofu. Whether you're just starting out with IaC or looking to solidify your knowledge of OpenTofu, this workshop is for you.
At Massdriver.cloud, we are always keen to bring you the latest insights and thought leadership in the world of platform engineering and DevOps. Our latest podcast episode is a must-listen for anyone looking to deepen their understanding of cloud infrastructure and its evolution.
ClickOps got you down? In this post we'll learn how to generate OpenTofu code using AI. Its not a silver bullet, but it'll get you further than many of the tools in the OSS space.
Enhance your Massdriver experience with our latest graph updates! Enjoy faster, more intuitive navigation with new features like line icons for easy connection deletions, a minimap for quick access to complex infrastructures, and persistent user-defined layouts. Simplify environment comparisons with a single click. Plus, check out our Platform Engineering Podcast featuring Mitchell Hashimoto and streamline your procurement through AWS and Azure marketplaces for seamless billing and deployment.
Why Terraform's New Test Functionality is a Step in the Wrong Direction
For teams ingrained in DevOps practices, this term ushers in a broader horizon, a fresh perspective on managing infrastructure. But what does Platform Engineering offer, especially for those adept in DevOps?
Generating OpenTofu or Terraform from existing cloud resources is a tedious process fraught with gotchas. It's not a silver bullet for click-ops debt. This post will discuss caveats of tools and different approaches to generating IaC.
Improving your organization's position in the DevOps Maturity Model involves focusing on specific practices, tools, and processes that align with the capabilities and needs of each maturity level.
Using common DevOps tooling shouldn't be this hard.
A misaligned text bug that was not CSS.
Observability in K8s With Jaeger
Massdriver's infrastructure bill of materials enables the feedback loop that powers cost and monitoring. Harness that power without transfering mangement control.
The Massdriver bundle spec helps engineers provision applications and infrastructure in the cloud while shrinking the surface area of the cloud API
Massdriver simplifies cloud monitoring with an intuitive graph of in-context metrics, allowing for quick creation of dashboards, alarm setting, and efficient system troubleshooting.
Managing cloud costs is a difficult task. Massdriver makes it easier with cost insights.
Infrastructure-as-Code has a pivotal role in bolstering disaster recovery and business continuity is substantial.
Massdriver Remains Unaffected by HashiCorp's Terraform Licensing Changes
Announcing Our $8m Series Seed Led by Builders VC
How HashiCorp's BUSL Change is Actively Hurting the Open-Source Community
Massdriver pledges support for the Future of Terraform: OpenTofu
Massdriver Changelog May 31st 2023
Massdriver Changelog April 6th 2023
A quickstart on running Kubeflow locally on Docker for desktop Kubernetes (part 1)
Dialyzer is a powerful static analysis tool for Elixir. Here's how to to use it in GitHub Actions with minimal impact on build times.
Massdriver Changelog March 8th 2023
The CIDR problem, and how we fixed it with Auto-CIDR
Massdriver Changelog January 19th, 2023
A Critique of How We've Fooled Ourselves for Years.