AI and Weak/No Code: What They Can and Can’t Do Together

Artificial intelligence (AI) is on the fast track and moving toward mainstream enterprise acceptance, but, at the same time, another technology is making its presence known: low-code and no-code programming. While these two initiatives occupy different spheres within the data stack, they nonetheless offer intriguing opportunities to work in tandem to dramatically simplify and streamline data processes and product development.

Low-code and no-code aim to simplify the creation of new applications and services, so much so that even non-programmers – that is, knowledge workers who actually use these applications – can create the tools they need to accomplish their own tasks. They work primarily by creating modular, interoperable functions that can be mixed and matched to meet a wide variety of needs. While this technology can be combined with AI to help guide development efforts, it’s unclear how productive the company’s workforce can become in a matter of years.

Smart programming

Venture capital is already starting to flow in this direction. A startup called Swing AI recently launched a drag-and-drop platform that uses open-source AI models to enable low-code and no-code development for novice, intermediate, and expert users. The company says this will allow organizations to bring new tools, including smart tools, into production more quickly, while fostering greater collaboration among users to develop and integrate these emerging data capabilities in a way that is both efficient and effective. highly productive. The company has already adapted its generic platform for specialized use cases in healthcare, supply chain management and other industries.

AI’s contribution to this process is basically the same as in other fields, says Gartner’s Jason Wong – i.e. taking on repetitive and rote tasks, which in development processes includes such things as performance testing, quality assurance and data analysis. Wong noted that while the use of AI in no-code and low-code development is still in its infancy, big hitters like Microsoft are keenly interested in applying it to areas such as analytics. platform, data anonymization and user interface development, which should significantly alleviate the current skills shortage that is preventing many initiatives from reaching production-ready status.

However, before we start dreaming of an optimized, AI-powered development chain, we’ll need to solve a few practical issues, according to the developer. Anouk Dutree. For one thing, abstracting code into composable modules creates a lot of overhead, which introduces latency into the process. AI is increasingly gravitating towards mobile and web applications, where even 100ms delays can drive users away. For back-office apps that tend to run quietly for hours on end, this shouldn’t be much of a problem, but it’s probably also not a ripe area for development with little to no code.

Constrained AI

Also, most low-code platforms aren’t very flexible, as they work with largely pre-built modules. AI use cases, however, are usually very specific and depend on the data available and how it is stored, packaged and processed. So in all likelihood you’ll need custom code to make an AI model work well with other elements of the low/no-code model, and that might end up costing more than the platform it is. -same. This same dichotomy also impacts functions such as training and maintenance, where the flexibility of AI clashes with the relative rigidity of low/no code.

However, adding a dose of machine learning to low-code and no-code platforms could help soften them and also add a much-needed dose of ethical behavior. Dattaraj Rao from Persistent Systems recently highlighted how ML can allow users to run pre-built models for processes such as feature engineering, data cleaning, model development and statistical comparison, all of which should help to create transparent, explainable and predictable models.

It’s probably an overstatement to say that AI and no/low-code are like chocolate and peanut butter, but there are solid reasons to expect that they can improve the forces of the another and reduce their weaknesses in a number of key applications. As business becomes increasingly reliant on the development of new products and services, the two technologies can remove many of the obstacles that currently stifle this process – and will likely remain so whether they work together or independently.