There are several things I like most about Semaphore. First, the UI is fantastic. I find it really easy to navigate and use. Integrating Semaphore with our GitHub account was super easy, and we’ve seen very few performance issues. When it comes to pricing, I think it’s excellent for what you get out of the platform. Semaphore also makes it easy to see which projects have the biggest spend, so you can adjust if needed and try to reduce costs. I recently opened a support ticket for an issue that turned out to be on GitHub’s side, but throughout the process the support agent kept me up to date at every step, which I really appreciated as a customer. Admittedly, I haven’t explored Semaphore’s AI offerings much yet, but I’m keen to see how AI can help us use Semaphore even more efficiently.

Close to the “metal,” it lets us express complex build and deployment patterns while still providing enough “sugar” to make common patterns faster and easier.

I like the Semaphore config file because it serves as a single source of truth. The pipeline lives in the repo alongside the code it tests, which is version controlled. I find making changes and updates via the Semaphore config file to be very straightforward.