A potential method for automating the conventional machine learning workflow is called auto machine learning (Auto-ML). The goal of auto-ML is to lessen the amount of manual work that must be done while developing, testing, and deploying machine learning models. The present Auto-ML approaches, their advantages, drawbacks, and potential future research paths are conceptually evaluated in this paper. We look at how Auto-ML may improve model accuracy, decrease the need for human intervention, and democratize machine learning. We also talk about the limitations, such as the difficulty in understanding Auto-ML models and the danger of relying too much on automatic methods. Additionally, draw attention to the Auto-ML research needs, such as the demand for more effective hyper-parameter optimization techniques and models that can be explained. Overall, this paper offers a thorough analysis of Auto-ML methods and acts as a guide for further investigation in this field.