Overview of AI application development environment

I would like to present an overview of the environment for developing AI applications. Look at the diagram below that I created. Basically, we can develop apps in traditional programming languages. However, in recent years, a variety of tools, libraries, and frameworks are on top of it and can be developed at a much higher level. Furthermore, AI services on the cloud and APIs that can be called in various languages ​​are available. High-performance hardware is essential for performing AI processing. Deep learning takes a lot of work to prepare the required datasets, but some systems ease it. Don't miss out on the transfer learning for efficient training using already trained models. The need to run AI apps on smartphones is also growing. Block programming environments such as MIT App Inventor support this.