In order to build a machine learning system, it is advised to start with a new small project and improve it progressively:

1. Find a similar problem to yours and download code (and test the model to check results)
2. Find ways to scale your computation if needed (namely, AWS/Google Cloud)
3. Start with smaller datasets to avoid losing time just waiting for a single epoch
4. Start with a simple architecture
5. Use visualization/debugging (for instance, TensorBoard)
6. Fine-tune the model, fine-tune hyperparameters, depth, architecture, layers, and the loss function
7. Expand your dataset and ensure that it is as clean as possible
8. Split your dataset into training, development, and testing sets
9. Evaluate your model