How are you automating your feature engineering? Spending more time on this than the model itself.

I’m stuck in a loop where I'm constantly building and rebuilding the same feature pipelines for every new project. It's incredibly time-consuming and feels like I'm reinventing the wheel. Are there any tools or libraries that help automate and standardize this process, especially for tabular data? Something that integrates with a standard sklearn/pyTorch workflow would be ideal.