Interview with ServiceNow: ‘Low-Code & GenAI Will Reduce the Developer Shortage’

What are we lacking right now? Skilled developers, with research from market researcher IDC, with a predicted possible shortage of four million developers over the next few years.

Other sources go higher, suggesting the developer shortage is 40 million globally and that the overall impact of slower software development on the economy could be worth hundreds of billions of dollars in years to come.

Whether you consider the lower-end or higher-end estimates more likely, we are hitting a wall in terms of what the world needs from software development due to a shortage of skilled developers to deliver it.

There may be two interlinked emerging technologies that can at least lessen the demand — the rise of low- and no-code platforms (which allows non-developers to cook up software) and generative AI stepping in to guide and write code.

It’s an opinion shared by Nvidia CEO, Jensen, who believes the next generation should not focus on coding, as AI will take over the majority of the work (YouTube).

We sat down with Jithin Bhasker, General Manager and Vice President for the App Engine business at ServiceNow, to figure out how much no-code may impact a stretched, capacity-filled industry.

About Jithin Bhasker

Jithin Bhasker is the general manager and vice president of the App Engine business at ServiceNow.

Over the course of his career, he’s led various initiatives to build IT/developer products and tools and has led strategic mergers and acquisitions. Prior to joining ServiceNow, Jithin worked at Google, Adobe, Dell, and other B2B startups, leading teams in product, corporate strategy, and engineering.

Challenges of the Developer World Today

Q:  The developer shortage continues to challenge businesses. How do you see this impact in the IT landscape, particularly in things like project delivery and innovation in the workplace?

A: As leaders, we’ve all struggled with the shortage of engineers, and the talent shortage continues to be a problem. We are seeing a huge demand for highly skilled engineers with the skills to fine-tune the custom-built large language models (LLMs) that are the engine for everything artificial intelligence (AI) uses at the back end.

According to the US Department of Labor, software engineering is among the top four highly in-demand roles and skills.

It’s projected that over the next ten years, there will be a need for at least half a million developers and engineers just in the US to support the growth in demand.

This is without considering the amount of augmentation and enablement that generative AI would do.

I foresee these numbers doubling in terms of the market for talent who could effectively drive this fine-tuning and refining of some of these models with LLMs all in place as well.

Q: Citizen development is another big buzzword at the moment. Can you explain the concept’s significance in today’s tech-driven business environment?

A: A recent economist survey stated that every company has at least 6 to 12 months of project backlogs in IT on average. Once again, that’s primarily because of the proper skill set and talent shortage. As a result, IT leaders and CIOs recognize the need to delegate this development work to the edge of the operations and the business processes.

The best people who could effectively automate, build applications, and custom-build all these digitization processes are at the forefront of those operations. You could be in sales, marketing, legal, finance, or anywhere.

But every CIO is looking at how we could delegate these opportunities to find ways to make the processes automated, digitized, and faster with the help of platforms like low-code and no-code platforms. It’s now driving the boundaries with the help of GenAI.

When you combine the two, that’s the combination of GenAI and the low-code, you’re enabling anyone to build, automate and digitize processes. The only limitation is your imagination or the ability to identify the actual pain point or the bottleneck within the process or the business operations.

So, now you are giving that ownership directly to someone at the edge of those business processes rather than highly depending on an IT or a developer.

Upskilling Knowledge in a World of New Tools

Q: What skills will be more crucial for professionals to train, monitor, and optimize new AI solutions effectively?

A: For a seasoned developer, brand new language models are introduced every other month or so. For example, after the buzz of OpenAI and ChatGPT, we saw Llama getting so much more prevalent, and now we are seeing models like Mistral getting so much more prevalent.

So, as a seasoned developer, it’ll be great for you to fine-tune or tune into all this rapid emergence happening in the market so that you are keeping yourself updated about some of the advantages and efficiencies that are coming up with this model.

Our CIOs or IT teams have started leveraging things like text-to-code within our company. They’re immediately seeing an efficiency of 30% more. Imagine you are saving that 30% capacity, which goes back to my earlier talk track about the shortage of developers.

As and when these models are getting more and more fine-tuned and more imaginative, you see a possibility of a seasoned developer being faster multiple X times in terms of designing, developing, deploying, testing, and deploying the code in a much better fashion.

Fundamentally, it’s about making seasoned developers multiple times faster to bring all this together, from design to development to testing to deployment.

But for someone who is not a skilled coder, you’re now reducing the entry barrier for the person to be able to get started with building things rather than in the old way where you are first trying to define the process.

Then you transfer that to an IT person to look at all these different systems, which they need to integrate and stitch together components that need to be brought together.

We are even looking at things like “text to test,” where you will run much faster cycles for your automation testing before you deploy into production. We see efficiencies across the different skill sets of developers, whether you’re new to coding or you’re a seasoned developer.

The ‘Democratization’ of Coding Skills

Q: How do you think the democratization of development will impact the industry?

A: It goes back to whether IT and the CIOs genuinely recognize the power of a low-code, no-code, and GenAI. Many of our enterprises and customers are learning and testing how to leverage this combination to place these responsibilities and opportunities into the business functions themselves.

The phase every CIO and the enterprise technology teams are going through is testing it by themselves. Once that phase is over and they have enough confidence, they will start delegating and distributing that capability and the skill set of those platforms to the business functions.

This could disrupt and transform the way every enterprise drives business processes or innovations and will have a tremendous impact on time to market or the efficiencies they would gain daily from those business processes in the future.

The backlog will likely come down, and more business functions will take ownership to build automation and applications. This will make it efficient for them to run daily operations and business. All of which make them more agile, easily scalable, and [a quicker] time to market; all of those benefits would come as a part.

Q:  How do you see generative AI reshaping traditional app development paradigms and generating business value? 

If you step back and decouple those words of low-code and no-code, generative AI would now make a low-code an absolute no-code.

Imagine you are someone in the marketing team who wants to build or automate the process of event tracking as a marketing team. You can now use simple, natural English language to “build me an events tracker.”

We already have capabilities like text, where simple English statements will generate the snippets of code for you, and text flow, where more or less you can describe the process flow. Our platform will also generate those workflows for you on the fly in a few minutes.

Where ServiceNow Fits In

Q: How are you positioning yourself in citizen development, and what kind of unique offerings does it bring?

A: ServiceNow was started in 2004, and the original mission or the premise for which our founder, Fred Luddy, started the company was anyone should be able to automate and build business processes on our platform. So automation as a buzzword isn’t anything new for us; it’s been embedded in the core of our platform from the first days of our inception.

Now, as a company, as a platform, and as the product portfolio that I own as an App Engine, it is all about the low-code, no-code platform, which you can now get access to where you can build any business processes and digitize this business processes in a much more accessible, more straightforward fashion.

Our App Engine Management Center allows teams to delegate to anyone who can build and automate those processes.

It also provides the CIO and IT teams with the proper control and governance of who is getting access to what data and who can build what, all the way from access control and role-based access controls fundamentally very much built into the technology and the platform itself. That’s why we are bringing it all together.

Q:  How do you envision the role of citizen developers evolving, and what are the key challenges and opportunities that you see happening in this space?

A: By default, delegated development will take precedence, and I’m using the words delegated development and citizen development interchangeably. It will take precedence, and that’s the only way companies and enterprises can truly scale the efficiencies and the velocity they have to keep up when in a hyper-growth mode.

However, it’s essential that you also have the proper guardrails and governance in place because now you’ve given such a powerful technology to many people within the company.

People will be empowered to build process automation integrated with the correct data set and connect multiple different systems or various business operations and processes quickly with the help of these two powerful technologies.

So you need to ensure you are giving the proper data controls, privacy controls, and access limits on top of all these things.

While you get familiar with this delegated development model and testing, you also need to ensure that you’re putting the proper governance and guardrails in place in the company.

Only then should you expand this platform and generative AI capabilities across the more extensive section of your employees within the company.