As generative AI continues to develop, many companies are wondering how the evolving technology can support their software development. Adoption has thus far been slow, with most businesses implementing AI only for major development projects.
“It’s a mixed bag,” said Aju Mathew, Vice President of Software Engineering for Aspire Systems. Some companies are starting to use AI for small proof-of-concept projects, while the more adventurous firms subscribe to generative AI platforms and provide basic guidelines for employee usage.
However, as companies get used to generative AI and the technology continues to evolve, Mathew expects AI’s role in creating software to be enormous. In this extended interview, he gives his take on current trends and practices in AI-based software development as well his view on the future of AI in creating software.
Watch the full interview or jump to select interview highlights below.
One of the more modest ways that firms use artificial intelligence in building apps is getting feedback and insights from generative AI programs, Mathew said. These insights include competitive analysis and market positioning for the potential new software.
He believes that, looking ahead to when companies use AI at a higher level, they’ll “allow generative AI to generate the application architecture and the deployment architecture based on technical inputs and parameters.” Clearly, this level of AI-based development won’t be easy to attain. “I’m sure there will be a lot of tweaks that will be needed, but it’s definitely possible and doable,” Mathew said.
Boosting this process is an emerging technique called prompt chaining. Instead of entering single prompts, developers are providing a chain of prompts to the generative AI tool, speeding the process.
Eventually, developers may create complete entire “epics” using AI. In software development, an epic is a large body of development work that can be divided up into smaller elements.
“Companies will allow the generative AI platform to consume the business requirements document—or even the recordings of requirements discussions with business stakeholders—and with the right type of prompt chaining and some advanced techniques, we can envision generating the entire epic.”
(These comments have been edited for length and clarity.)
For more information about generative AI providers, read our in-depth guide: Generative AI Companies: Top 20 Leaders
The post Aspire’s Aju Mathew on Generative AI in Software Development appeared first on eWEEK.
