Automated accurate segmentation of nuclei on bioimage is one if not the most often required task in image analysis project in core facilities and research laboratories yet it is still a complex and a time-consuming task with classical approaches. The advent of new deep learning-based tools that are easy to use by biologists without any Deep Learning (DL) expertise has been a game changer. We will use out-of-the-box StarDist as an example tool to guide beginners through their first use of their first DL cell nuclei segmentation using open source software Fiji and QuPath. The workshop will present 1) an overview of the limitations of classical segmentation approaches and the advantages of StarDist for this task, 2) demonstrate how to apply the DL tool to the segmentation tasks and 3) most importantly how to evaluate the results and identify current limits to the out-of-the-box DL approach.