Low code and no code may open more doors to artificial intelligence

The jury is still out on whether low and no code platforms can blaze a path to high-end application development -- at least not yet.
Written by Joe McKendrick, Contributing Writer

How high will low code applications go? The jury is still out on how high this is all going. Low code and no code may even play a role in enabling business users to build artificial intelligence-driven applications, some observers predict.

Photo: Joe McKendrick

Low and no code platforms make it possible "to deploy artificial intelligence without hiring an army of expensive developers and data scientists.," states Jonathon Reilly, writing in Harvard Business Review. "Removing friction from adoption will help unleash the power of AI across all industries and allow non-specialists to literally predict the future. In time, no-code AI platforms will be as ubiquitous as word-processing or spreadsheet software is today," 

Reilly advises looking for platforms that make such development as easy as possible -- a simple interface that integrates with popular enterprise applications; that data is automatically classified; automates model selection and training; and monitors model performance. "The user should not need to know their way around regression or k-nearest neighbor algorithms."  

Not everyone agrees that low code and no code platforms can blaze a path to high-end application development -- at least not yet. "I don't believe the average business user can build and deploy sophisticated applications," says Kevin Harbauer, chief technology officer at Ephesoft. "Most of these low-code/no-code solutions require a fairly technical power user to be successful." 

The art of creating a great application of any complexity "still requires work around user experience design, prioritizing the right features, and understanding workflows and algorithms," says Ed Sawma, VP of operations and marketing at Transposit. "it's only the implementation that's moved to drag-and-drop interfaces that don't require writing code."

Low-level automation is a popular use case, which "seems to be where the bulk of where these types of applications are being used outside of IT," says Harbauer. "These solutions allow power users to automate simple tasks, integrate data between systems and develop simple applications that would have required IT in the past."

This goes to the heart of what it means to be a citizen developer. "The goal with low-code or no-code software is not to have citizen developers build sophisticated applications or mission-critical software applications," says Harel Tayeb, CEO of Kryon Systems. Rather, it is intended to enable citizen developers "to quickly create an app that will help them solve a problem, and then move on with their workday. More sophisticated applications will still require IT or engineers to accomplish."  

Robotic process automation, for one, "is still too complex for the citizen developer," Tayeb says. "Vendors still have a lot of work to do. In order for citizen developer-based RPA to really take off, RPA tools must be truly easy for users, meaning no coding experience or technical background needed. The tools must be intuitive and empower the user with the features needed to design and deploy a software bot."

Where the citizen developer can add value "is that they know the process that they want to automate, inside and out, because they have been performing that process manually," says Tayeb. "A short learning curve and minimal training is imperative. Some platforms already provide this. and some are still working on it."  

With growling sophistication comes growing problems. "There's a risk that citizen developers might spend significant time building something in a no-code-only platform that then needs to be re-coded from scratch to actually be fully functional and usable by an organization," says Sawma. "To prevent this scenario, team members should communicate openly about what is being built in their no-code platforms so that time and resources aren't unintentionally wasted." 

There is also a rub if the level of sophostication with citizen developer-built software does increase to advanced analytics or AI levels -- who will maintain or update these apps? "As we saw with the rise and fall of Microsoft Access, the biggest risk to organizations is that once a business user moves on to another role or leaves the organization, who will maintain all the business-critical applications they built?" Harbauer points out. "Just like traditional software development and enterprise application management, strong change management and enterprise application oversight is critical with these one-off applications built by business users."

Editorial standards