I'm currently exploring options for infrastructure management as I start a new project, and I'm particularly interested in Pulumi. However, I'm wondering about the long-term viability and whether investing time in learning and using Pulumi is worth it compared to more established options like OpenTofu, Terraform, or Ansible. What do you all think?
5 Answers
I suggest sticking with OpenTofu and Terraform for infrastructure tasks due to their maturity. Ansible and Packer are great for server configurations too. It’s about keeping a clear desired state and managing resources effectively.
I use OpenTofu for all my projects, mainly because I find it reliable and easier to maintain. I think if you want a straightforward infrastructure management experience, sticking to a declarative approach is the way to go.
For me, it's all about the declarative model. Using a consistent syntax is easier for most team members from varying backgrounds. I really believe that programming-based solutions add unnecessary complexity for simply setting up infrastructure.
While I get what you’re saying, I have to argue that Pulumi is also declarative. It allows you to specify the desired state and manages the process to achieve it. Plus, they've introduced support for YAML and plan to give more options soon.
I disagree; I think keeping it simple without a programming language approach helps avoid issues that arise from overly complex setups.
We chose Pulumi because it integrates well with TypeScript, our core product language. So far, it’s been a solid decision as we transition from legacy Terraform setups. If you have questions, I’m here to help out!
I'm not a fan of coding for infrastructure. I think declarative languages do the job better and help maintain clarity.
Pulumi is definitely appealing if you prefer to write real code instead of using HCL, but keep in mind that it’s a smaller company compared to the established giants like Terraform. Terraform has the advantage of widespread recognition and support from DevOps professionals, which can be a significant factor. If your team is familiar with Python or Go and not a fan of YAML, then Pulumi could be great for you. Otherwise, it might complicate onboarding for new hires who are used to Terraform.
That's an interesting take, but Pulumi isn't as small anymore. With changes happening in the industry, like IBM's acquisition of HashiCorp, big companies can kill products unexpectedly. We have to be cautious about what's upcoming and how tools will evolve.
I feel you on this point. It’s 2026 and AI is changing the game; writing infrastructure code by hand is becoming less common. I believe the flexibility Pulumi offers with programming languages gives it an edge.
Actually, I work at Pulumi and I can clarify that Pulumi operates on a declarative basis too; it lets you declare your resource states and automates the rest.