Paper Type
ERF
Abstract
The design of LLM based apps is like attending Mardi Gras. The outcome is unpredictable, but the results are often amazing. The design of AI systems needs a fundamentally new paradigm for the development of those systems because of their unprecedented call or innovating into novel solutions with competing solutions. Although the traditional systems development life cycle may include alternative analysis in its early phases, it did not consider fundamentally different approaches. Scrum attempted to improve upon SDLC for novel environments through iterative development that could adjust based on the results of early Sprints. With generative AI, the results can be surprising, and the providers are quickly changing calling for greater use of experiments to understand what is possible and in a solution. This paper describes a software development methodology for developing generative AI systems that embraces experimentation to produce better results.
Paper Number
1767
Recommended Citation
Walsh, Kenneth R., "Developing Generative AI Applications" (2025). AMCIS 2025 Proceedings. 2.
https://aisel.aisnet.org/amcis2025/sig_sand/sig_sand/2
Developing Generative AI Applications
The design of LLM based apps is like attending Mardi Gras. The outcome is unpredictable, but the results are often amazing. The design of AI systems needs a fundamentally new paradigm for the development of those systems because of their unprecedented call or innovating into novel solutions with competing solutions. Although the traditional systems development life cycle may include alternative analysis in its early phases, it did not consider fundamentally different approaches. Scrum attempted to improve upon SDLC for novel environments through iterative development that could adjust based on the results of early Sprints. With generative AI, the results can be surprising, and the providers are quickly changing calling for greater use of experiments to understand what is possible and in a solution. This paper describes a software development methodology for developing generative AI systems that embraces experimentation to produce better results.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
Comments
SIGSAND