What is our primary use case?
My main use case for Pulumi Enterprise is to deploy resources to our main project called MIA, which is basically an AI assistant. We use resources from AWS, including S3, EC2, ECS, ECR, Lambda, SNS, SQS, and others.
For example, we have a notifier service where we have a Lambda worker that sends messages to an SQS queue, and with this, we have a consumer that sends WhatsApp messages and SMS messages to our customers.
We are using Pulumi Enterprise with our AI assistant, which is Cloud Code. We encountered some difficulties integrating this with our code assistant, so I believe it could be easier to integrate. I remember a time when Cloud Code made a mistake in the documentation, creating a non-existent parameter and props, which is a common issue I see during my workflow.
What is most valuable?
Pulumi Enterprise offers very good features. I really appreciate the access control with permissions and detailed permissions by role or team, which we use for our developer team. This is the main feature I appreciate, as there are different levels within our team. Policy management, policy as code, and audit logs are also beneficial because when a starter developer makes a mistake, we can check the audit log.
Pulumi Enterprise has positively impacted our organization by enabling fast deployments and providing a centralized configuration for our two environments, development and production. With separated stacks, we can replicate the same environment on both sides.
Before using Pulumi Enterprise, our deployment time was between thirty minutes to an hour. Now with Pulumi Enterprise, we can change anything in our cloud environment in just a few minutes, like two or three minutes. Of course, there are more complex situations that differ in time, but in general, we have reduced time by about sixty to seventy percent.
What needs improvement?
The main feature that could be improved would be better integration with AI. For instance, having an assistant in the platform, as a chat, to ask questions and make changes with this conversational model would be beneficial.
The main improvement for Pulumi Enterprise could be the integration with AI, allowing it to learn and make changes to accounts or stacks. My main concern is really the AI integration. Regarding documentation, it could be interesting to have a markdown copy button. When I go to S3 documentation or Lambda documentation, it would be great to have a copy as markdown button, as I have seen in other documentations, which is very good for providing context to AI.
Regarding Pulumi Enterprise's AI capabilities, we currently have good control over security and governance, but it is difficult to give control to AI. There is a good approach from AWS where the AI functions more as a supporter rather than a manager, acting as a consultant where you can get information and ask questions without losing control or granting high permissions to AI.
In the past year, I remember instances when Cloud Code searched for documentation from Pulumi and usually returned a non-existent parameter, which caused me to spend a lot of time finding and removing the incorrect parameter. I think it could be easier for AI to crawl documentation pages and handle such tasks.
For how long have I used the solution?
I have been using Pulumi Enterprise for two years, two and a half to be precise.
What do I think about the stability of the solution?
Pulumi Enterprise is stable, and we have never encountered any stability issues.
What do I think about the scalability of the solution?
Pulumi Enterprise offers good scalability because we can add more members and create new stacks, migrating with ease, which is beneficial for us.
Pulumi Enterprise is a very good solution for scalable teams, and we are trying to get better with this.
How are customer service and support?
To be honest, I have never required specific support, but the main, first-level support has been very good.
Which solution did I use previously and why did I switch?
We used to deploy and make infrastructure and cloud changes manually. Our process before Pulumi Enterprise was very hard and slow.
What was our ROI?
We have seen a return on investment because it is easier to make infrastructure changes, even without designated DevOps personnel. Developers can make these changes, and we save time due to these automations, allowing us to change environments in less time than we could two years ago.
What's my experience with pricing, setup cost, and licensing?
My experience regarding pricing, setup cost, and licensing has been good, without any problems. We find the situation stable and plan to continue using Pulumi Enterprise for many years.
Which other solutions did I evaluate?
I evaluated Terraform, but we chose Pulumi Enterprise as the option for us because of the programming language. We prefer using Python since our main stack here in our project is based on Python, which most developers work with, while Terraform uses a different language.
What other advice do I have?
I would advise that Pulumi Enterprise is a good option, but it is important to study and gain context before asking AI questions. It is crucial to understand what you are doing in the cloud environment, particularly regarding AWS or other providers. I would rate my overall experience with Pulumi Enterprise an eight out of ten.
Which deployment model are you using for this solution?
Private Cloud
If public cloud, private cloud, or hybrid cloud, which cloud provider do you use?
Amazon Web Services (AWS)