About Consistent pipelines, IaC codes and IA Genetative with Stable Diffusion (Part 1)

Software is a combination of code and documents. That is, with recommendations, it means having standards and agreement on them with standards of good practice. What determines the programmer's written code pattern? Among different thoughts about this, in fact, in a simple way, the programmer declares a variable by abbreviating or shortening or increasing its name, or more than anything else about it, according to the increase in code in the systems, which is why it could be some cause of failure and errors such as variable declaration.

All actions regarding the use of any programming language using its specific functions and covenants promote a mentality of legal or illegal standards that will define clean codes. These patterns, when well defined and evaluated, are classified as weakness, bad smell, reuse, versioning, increase and productivity. In Software Engineering, using the power of DevOps concepts and Infrastructure as Code (IaC) tools such as Puppet, Salt, Ansible, among others.

Thus, these tools promote scripts from programmers around the world in open source code repositories such as Github, Gitlab, Bitbucket, others (CI practices - Continuous Integration, CD - Continuous Delivery) who still strive to ensure good provisioning of their codes and prevent these resources from failing, for example: an account credit payment service that needs to be implemented with a resource called CALCULATE_WITH_TAX_PRICE at any future time, executing process steps in a pipeline in the Cloud... It is the execution process in different stages and this is evaluated on points such as quality, semantic code, quantity of code (LOC) and mainly the operation.

If they fail the provisioning steps, the security or DevOps team will receive failure messages (LOGs) about this. Thus, the team's mission is to discover and understand all possible causes of failures, such as security flaws, vulnerabilities, bad smells, network failures, invalid parameters, and any others, in this complex and slow journey.

Performing a new run and fixing problems in the pipeline becomes very expensive in addition to other problems that have not yet been identified. Furthermore, it is necessary to understand the use case of the running system associated with possible wrong code, lack of standards, semantic and cohesive code that the developer wrote, even if he has the best intention about the concept of the product or feature.

In my research, I am developing a tool that has concepts of generative AI and stable diffusion. I will test this tool a lot with experiments and hope that it will help all developers in the community with the best code day after day.