Bring your towel and join one of the fastest growing data communities. Welcome to our second-generation open source orchestration platform, a completely rethought approach to dataflow automation.

Prefect Community

Hi, I have seen Prefect does not use pipenv nor poetry. How do you virtualize the environments when working directly with pip? Do you use virtualenv to create multiple python distributions and then work with plain pip?

all options are valid and depend on your personal preference - Prefect should work with all of those options

I personally use Conda and find this is the easiest, and you can even supply the conda environment when deploying flows with Prefect 2.0:
```DeploymentSpec(
    name="dev",
    flow=test_flow,
    flow_runner=SubprocessFlowRunner(condaenv="yourvenv"),
)```

but if you are asking how we develop and test, this is usually handled by CI with GitHub Actions - there is a matrix of various environments and we then test on Ubuntu across all Python environments in the matrix, configured with Action installing Python

I broke Windows build but works on other platforms

Windows is a bit complicated, we are working on it though

<https://github.com/PrefectHQ/prefect/pull/5958>