In there, some important projects built using ML and Deep Learning ANN.
Most of the dataset that I used to build the model is from kaggle. In there, I used ML algorithms and evaluate the performance. After that, I used Deep Learning to check the performance. I usually used EDA to find out the exact pattern of the dataset as I was most of the time ignorant of domain knowledge of dataset and performed Feature engineering to handle the dataset. Eventually, when I feel like that I have found the appropriate output that I wanted then I didn't go further investigation.