What is our primary use case?
We use GitHub Actions for machine learning and predictive analytic use cases, but not in generative AI, as we use it a lot in MLOps, for example, for the purpose of model packaging and automating model deployment. We use CI/CD from GitHub Actions as well.
How has it helped my organization?
GitHub Actions helps a lot with integration and also with automation, which eases our development processes.
It saves a lot of time with GitHub Actions; especially, the whole process if done manually might cost us a lot, maybe a week or two weeks just for deployment, but now with automation, we automate the deployment effectively.
The main saving with GitHub Actions is that we do not have to put our senior engineer for deployment.
GitHub Actions is cost-effective, especially when we consider doing it manually with an engineer, as that is very costly to us.
What is most valuable?
GitHub Actions' primary use case is CI/CD, and we use it for infrastructure as code. In one of our projects, we use Terraform to define the infrastructure, and we use CI/CD to make architectural changes to the enterprise infrastructure, and it works effectively.
The most helpful part of GitHub Actions is the automation; one is automation of infrastructure, and the other thing is that because we work a lot with data engineering, we do a lot of custom Python packaging. GitHub Actions has been very useful for that; for example, we can make code changes, and then the library is automatically built and published to the PyPI index.
What needs improvement?
Frankly, I cannot imagine something that could be improved in GitHub Actions; there's a lot of capabilities, and the feature set is more advanced than we use.
The pricing is high for the advanced security; GitHub Advanced Security for us is quite pricey.
For how long have I used the solution?
I have been working with GitHub Actions for quite a long time, maybe three to four years at least, especially on some recent enterprise projects.
What was my experience with deployment of the solution?
GitHub Actions is not difficult to initially set up and deploy. The UI is very intuitive, and we've set it up for many use cases. The GitHub Actions build agent setup for on-prem deployment was quite smooth and easy to set up.
We can set up a new project and the actions in just hours.
What do I think about the scalability of the solution?
GitHub Actions is scalable; we have no issues, and currently, I'm using it in a Telco setup, so for our Telco company, it is useful, and it scales effectively.
How are customer service and support?
We haven't needed much regarding GitHub Actions support. Recently since we started using AI Chat GPT for GitHub Actions, it has made our development much easier.
I don't require technical support from GitHub directly as I can handle it myself.
How would you rate customer service and support?
What other advice do I have?
The syntax for the action itself requires a bit of learning; the syntax and the language being used for composing the GitHub Actions is the only challenging aspect.
On a scale of 1-10, I rate GitHub Actions a 9 out of 10.
Which deployment model are you using for this solution?
On-premises
If public cloud, private cloud, or hybrid cloud, which cloud provider do you use?
Other