Story Planning For Iterative Software Generation

A product story comprises the roles and responsibilities of various doers and teams. Therefore it is ideal to have those teams and user experience representatives build their respective story. In a classical software development project, the SRS(software requirement specification) is compiled similarly, but the details are often blurred because of various levels of sub-documentation. Namely, a SRS is converted to some form of FS(functional specification), SA(system architecture), and other such documents suited for each aspect of the software development process. 

With FastBuilder, you only need to develop the product story and the rest is managed by the AI.

Story Subsection

The storyboard can be divided in multiple subsections. And each team can be assigned their respective sections for development.

Common Terminologies

In AI, words have It is essential to use common set of words and phrases for AAAMP elements. The teams should share the AAAMP element list through joint sessions. It would be ideal to promote common terms in daily communication as well.

How to be agile with Story?

Although, the software can be built instantly using FastBuilder, but testing and validation of meeting all requirements takes time. Therefore it is ideal to break the story development in stages so that each step of the development can be tested and validated before moving on to the next stage. The project can follow any frequency for staging depending on their own testing capabilities. If one can test a set number of requirements at the daily frequency, they can choose 1 day staging speed. Similarly, depending on the team’s capacity, the sprint duration can be decided. For 1-day sprint, teams should develop a part of the story at the beginning of the day. After that story board can be updated by an assignee. Next, the assignee runs the development process and delivers the new generated code for testing/staging DevOps or internal hosting. By afternoon, the testing team should start validating the features. Finally, based on the testing conformance, the next stage of story development can be decided. It may happen that at this stage, the project team can decide to take one or more of possible actions:

1. Improve story (one or more of the sections)
2. Improve the prompt-sets
3. Improve the AI
4. Improve the code manually 

Promptset Improvement

AI based code generation and software generation is in its early stages. Although there are huge gains of adoption but there are also considerable effort in ensuring that an organization or group is able to fully realize these gains. Promptset improvement is key task for accuracy of software generation. Depending on the story elements, the promptset might need some tweaking and editions. Refer to the promptset section to learn more.

Next: Story Session