Docker works on Mac, Linux and Windows, allows to spin up containers locally as well as use similar config for deployed instances. Great documentation and support. Docker desktop app is available with great UI that helps review and manage containers, get to the logs and observe the performance. Docker cli integrates great and allows CI/CD, scripting, etc. to automate the usage. From my experience docker's performance is best on linux system. When working with mac M-series you need to be careful and pick containers that work with M hardware. Free docker plan works fine for me, but the paid plans are not too expensive if you evaluate what you're getting. My current experience with claude code AI spec-driven development shows that claude tends to use docker compose a lot which works great especially for POC and local development.

It's great when we need some software and don't want to install it on our local machine. It spares us from having to configure the software to our machine requirements.

Honestly, what I love most about Docker is how it completely eliminates the “works on my machine” problem. Since I started using it about a year and a half ago for both local development and production deployments, my workflow has become far more predictable and consistent across environments. Being able to spin up isolated containers for different microservices is a real game-changer. I can run the entire stack locally without dependency conflicts, and deploying to production feels much less stressful because the environment is essentially the same as what I tested on my machine. One unexpected benefit I didn’t anticipate is how much faster onboarding new team members has become. Instead of spending half a day setting up a development environment, someone new can just pull the image and be up and running within minutes. That alone has saved us countless hours.