Hi Microwave! One Hot Code Please

TLTR. That’s the Gen Z and Gen Alpha’s gripe with everything we boomers and Millennials grew up with. From letters, books, designs, blueprints, workflows, spreadsheets, story-plots to cars, conversations, dinners, itineraries, diaries and photo-albums – a lot has become ‘Too Long To Read’ and understandably so—for the time and age the digital natives have been raised in. It’s normal today to just sit in a car and cruise—and not have to change gears at all. It would soon be normal to simply tell the car where you want to go and take a long nap while it gets you to the destination. The gear box and the intricate dance of machinery inside the hood – both- have vanished. To be replaced with chic and ultra-simple digital dashboards. Your job is just to think of what music you are in the mood for. Driving is, also and, finally ‘TLTR’.

And yet, there are car connoisseurs who love a car for its specs—those of the hard-core kind. There are photographers who still lean towards cameras that offer some complexity. There are still planes and spaceships where cockpits retain the room for human eyes and fingers. There are still luxury watches that are expensive because they run on old mechanical parts. Expertise and complexity are not bugs but features – in many areas.

Gartner predicts that by 2028, 30 pc of business role job descriptions will require low-code or no-code skills.

- Jason Wong, Distinguished VP Analyst, Gartner

So why does that not hold true for coding? Or does it? Especially when business users do not need developers on the steering wheel anymore when they want to go to an app destination themselves. Even developers are not finding complex gear boxes on their seats anymore. It has all just become so swift, so sleek, so Gen Z today. The question is: Is it worth the ‘skip button’?

Less Code, More Fun, More Speed—Because

Low-code and no-code approaches are a fantastic idea today because they significantly accelerate the development process, especially no-code, which makes building solutions more accessible to non-developers, captures Jason Wong, Distinguished VP Analyst at Gartner. “This democratisation through citizen development empowers businesses to innovate rapidly and respond to market changes more efficiently. Gartner predicts that by 2028, 30 per cent of business role job descriptions will require low-code or no-code skills as non-IT teams increasingly develop applications and automate process workflows.”

Noro Chalise, Data Scientist and Founder, NoroInsight offers some real-world examples to sketch the picture better. “Addiko Bank used Appian’s low-code platform to automate SME loan approvals, reducing turnaround time by up to 50 per cent. U.S. Air Force (CON-IT) replaced over 100 legacy systems using low-code, resulting in estimated savings of more than US$80 million.

Aviva Insurance consolidated 22 internal systems into a single interface using Appian, improving agent response speed by nine times. Unilever enabled non-technical staff to build over 400 internal tools using Microsoft Power Platform, supporting a wide range of operational needs.

No-code development tools have gained popularity in recent years as an alternative to more complex low-code platforms, adds Wong. “The 2024 Gartner Digital Worker Survey shows that 49 per cent of non-IT respondents are customising and building tools from technology provided by IT or from what they acquire on their own, and increasingly, these are done using no-code.”

Low-code/no-code (LCNC) solutions shorten time-to-market from months to days, notes Ankit Sharma, Senior Director and Head - Solutions Engineering at Cyble. “According to a 2025 Forrester study, companies that used LCNC solutions reduced app development time by an average of 65 per cent. Rapid idea-to-product cycles are enabled by the use of built-in templates, drag-and-drop elements, and pre-integrated services.

Nagasanthosh Joysula, Co-founder of TableSprint draws an analogy between calculation and coding, stating that low-code/no-code approaches are similar to using calculators. “To the extent they are able to perform and serve the purpose, they can make the job easier and increase adoption. This is not new, but what is different today? It is the scale and speed at which low-code/no-code approaches are evolving to deliver the tasks that till recently required coding. This evolution is changing the economics underlying the application development life cycle.”

He explains ‘how’. “Current no code approaches do generate code through AI but can still stay as no-code platforms for users. This has exponentially increased the diversity of applications that can be built in a no-code manner, thus the spread of vibe coding as a popular approach. The recent rise and acquisition of Base44 by Wix is an example of how fast the changes and adoption are happening. At the same time, products such as Roo Code and Cursor are increasing the ease and speed of coding, making the traditional way of application building also faster.”

Before calculators, people relied on mental or paper math. Calculators eased cognitive load, empowering even those uncomfortable with numbers to handle tasks requiring computation.

- Nagasanthosh Joysula, Co-founder, TableSprint

Low-code platforms like Microsoft Power Platform, OutSystems, Appian, Mendix, and Retool are widely adopted by enterprises for internal tools, automation, and data-driven apps. And No-code tools such as Bubble, Glide, Webflow, Airtable, Zapier, and Softr are popular among startups and small teams for quickly launching MVPs, Chalise observes.

Less Code, More Users, More Legos—Because

Are these approaches bringing on-ground speed, democratisation, and prototyping ease though?

Wong answers in the affirmative. “They democratise development by allowing business users to create applications without deep technical knowledge. This is achieved through intuitive drag-and-drop interfaces and pre-built templates that simplify the process.”

Low-code/no-code (LCNC) platforms aren’t just Buzzworthy technologies, underlines Sharma. “They’re mature technologies that meet an urgent set of needs germinating in today’s digital economy. In 2025, as the gap of global developer talent still lingers (an estimated 4 million shortage of developers globally, according to Evans Data), LCNC solutions fill the gap between the technological ambitions and the availability of developer talent.”

These approaches also expand the power of non-coders and business users.

Low-code and no-code approaches empower organisations to build and deploy applications faster, reduce IT bottlenecks, and encourage active business user participation in automation initiatives, adds Prasanth Krishnan, Principal Sales Engineer, UiPath.

With the combination of low-code’s visual tools with code-level access for professional developers, there is enough flexibility while maintaining speed of deployment, Sharma points out. “At Cyble, we have seen low-code platforms reduce development time for threat intelligence dashboards between 50-70 per cent—enabling cyber security analysts to rollout new workflows quickly. The no-code use cases shine in departments — internal Apps, marketing dashboards, or hour workflows – however, scale and integration are weak points for no-code. Low-code is a better fit for critical and scalable enterprise use cases, while it may have a little more work to manage.”

Business users with domain knowledge can participate in solutions without relying on development teams through rapid testing and development, Sharma adds.

But would these tools work well with APIs, containerisation, and customisation projects?

Wong assures that most modern low-code and no-code platforms are designed to integrate seamlessly with APIs, and some support containerisation to be deployed on a cloud provider of the customer’s choice or even on-premises.

These platforms are excellent at containerisation, customisation, and API integration, Mark Weaser, Vice President at OutSystems Asia-Pacific also assures.

Less Code, More Vibe—Because

With the advent of AI, a new genre of less-coding has come on the surface. It is called ‘Vibe Coding’.

Vibe coding has reduced the number of steps required as part of application development life cycle, Joysula dissects. “There were days when customers used to show some references to technical teams, which was followed by requirements gathering, low-fidelity wireframes, high-fidelity wireframes, prototypes and so on. Now, the processes are becoming more collaborative. Business users are able to create the prototypes and first cuts of MVPs by themselves.”

With the emergence of AI-assisted LCNC tools (such as Copilot in Microsoft Power Platform and OutSystems AI Builder), business users can build apps by describing functionality in naturally spoken or written language, Sharma illustrates.

As Weaser assesses it, with the low-code/no-code approaches, software is being realised faster – a key advantage as IT leaders face mounting pressure to deliver business value from technology investments under tighter timelines and limited resources. AI-powered low-code platforms are empowering developers and business users alike through intuitive visual building tools and pre-packaged components to create applications up to 50 per cent faster than traditional coding-based implementations.

He also shines the light on some adjacent side-effects. “The dependency on specific programming languages is becoming less mandatory for new and stand-alone applications. This shift also brings the advantage of clear communication of requirements and goals increasing efficiencies in project management.”

Lightweight no-code platforms may not provide enough customization for advanced use cases, which often leads to rebuilding once the product needs to scale or integrate with more complex systems.

- Noro Chalise, Data Scientist and Founder, NoroInsight

Weaser adds that by enabling teams to rapidly build functional models, AI-powered low-code makes it easier to validate concepts early and gather stakeholder feedback. This ‘shift-left’ approach helps surface issues earlier in the development lifecycle, leading to higher-quality applications and faster time to delivery.

AI coding assistants like GitHub Copilot, ChatGPT, Replit Ghostwriter, and Amazon CodeWhisperer are now routinely used by development teams to boost productivity, cites Chalise.

But a lot of outcomes in this new age of coding hinge on good prompting skills.

Good prompting skills are valuable, but inherently there are limitations right now to what can be built just through prompting, Wong explains. “We expect a combination of visual and conversational approaches to be productive.”

There are many factors to consider, argues Sharma. “Like how well the user understood the data architecture, how clearly they communicated to the LCNC tool, and planning for the required logic to operate the app. In 2025, prompting well is becoming a new skill to develop, and when someone refers to ‘prompt literacy’ in relation to digital training, including technologies like Generative AI, many businesses have incorporated this into their training programs.”

Advanced prompting abilities are not a major factor in these platforms’ success—says Weaser in a contrarian argument. “The process is made simple for users with little technical expertise by drag-and-drop interfaces, guided workflows, and AI-assisted development. Although a robust understanding of business requirements improves results, built-in logic and templates lessen the need for specific prompts.”

So we know now that low/no/AI code gives speed, ease, fun and user-side context in an exciting way. Yet despite the surge and love being seen by these digital siblings of old-school coding, not every enterprise can find the courage to trust dashboards instead of developers. Let’s double-click on that.

Less Code, More Speed, More Users—But

One cannot turn one’s eyes and hands away from serious issues like lack of developer control and gaps in scalability.

There are also concerns that have been experienced with many low-code/no-code tools – specially on flexibility limitations, sloppy UI, inability to handle complexity, and security worries.

These worries are valid, especially in complicated, security-sensitive contexts like ours, Sharma assents. “No-code platforms, specifically, are usually unable to present system logic or just generate very long backend code, which is problematic for setting up a proper debug or audit. In 2025, Gartner reported that 42 per cent of organisations using a no-code platform ran into scalability issues when they crossed the mark of 10,000+ DAUs (Daily Active Users).”

Security remains a concern. Sensitive data on cloud no-code platforms is often poorly encrypted due to lack of granular access control. Weak integration with enterprise IAM can also result in Shadow IT.

- Ankit Sharma, Senior Director and Head - Solutions Engineering, Cyble

Another negative is the limited reusability or portability of the resulting no-code apps, Sharma tells. “In cases where an organisation decides to move entirely from one platform to another, say for example, moving off Salesforce as a means of application delivery and replacing it with Sharepoint, a total rebuild is likely. Low-code apps could be a slight improvement, particularly modern low-code apps in 2025 that back Docker Containers, Kubernetes, and REST APIs natively.”

These concerns are valid, particularly when these tools are isolated or not designed for enterprise use, concurs Krishnan. Many users report performance bottlenecks as workloads grow, Chalise agrees. “Built-in UI components can feel repetitive or too basic unless enhanced with custom development.”

It should not be assumed that these applications can be developed in isolation bypassing the IT teams, seconds Joysula.

Less Code, More Debt, More Shadows

There is also another concern that persists. It may get strengthened with digital tools. A new layer of Shadow IT and tech debt. What happens when there is a lot of rebuild work that low-code/AI-code work creates after the confetti settles down?

There is a risk of shadow IT and tech debt if governance is not enforced or if IT is too heavy-handed in what it allows the business to do, avers Wong. “There can be a careful balance of empowerment with governance. However, with proper oversight and integration strategies, these risks can be minimised. Rebuild work is generally less intensive compared to traditional development, but varies depending on the platform and application complexity.”

When it comes to tech debt, there are counter arguments too. There is a perspective that low-code tools could be good news for those worried about tech debt. “Businesses can allay these concerns by adopting a low-code platform with strong governance features. Features such as permission controls, audit trails, and compliance tools provide centralised oversight, helping to guarantee compliance with IT standards. Compared to traditional coding, where fragmented codebases can accumulate over time, low-code platforms minimise tech debt by simplifying maintenance and facilitating seamless updates, reducing the risk of unmanageable systems and promoting long-term maintainability,” Weaser explains.

Let’s zoom in on rebuild-work now. The amount of rebuild-work, as Chalise notes, depends on the platform and stage of growth. “Startups often use no-code tools like Bubble to quickly launch their MVPs. These tools are great for validating ideas and getting early traction. However, as the product grows in complexity, many teams find themselves rebuilding parts of the system using traditional code for better performance, scalability, and flexibility. On the other hand, enterprise-grade low-code platforms like OutSystems, Retool, and Mendix are built with developers in mind. They typically offer support for REST APIs, scripting capabilities, Docker, and CI/CD pipelines, which makes them more suitable for integration-heavy or containerised environments.”

There are more headaches and dilemmas that will pop up as this new genre of coding gathers steam. By 2025, platforms are evolving from mainly drag-and-drop into AI-generated code, comprised of Generative AI generated back-end logic or intelligent data schema modifications. This presents ethical and explainability challenges, highlights Sharma—how do we audit AI generated code? How do we know it is created to follow security best practices?

Due to their fair balance between speed and the level of customization and control required, low-code platforms cater more suitably to enterprise requirements.

- Mark Weaser, Vice President, OutSystems Asia-Pacific

Another significant, but not often discussed, concern that he reminds us of is that of vendor lock-in. “Many low-code/no-code platforms, for instance, utilise their company’s proprietary DSL (Domain-Specific Language) or component libraries. Organisations run the risk of becoming a highly dependent player for a long period without a way to export, or have standard interfaces to improve transition upon agency separation that turns cost, and limits flexibility to pivot on longer-term strategy.”

Another factor is that of oiling these wheels. Maintenance and ownership can become unclear when the original creators of apps or workflows leave the organisation. As Chalise reckons, some platforms make it difficult to export workflows, data models, or business logic, limiting flexibility in the long term. Deep integrations, complex workflows, or advanced business logic usually require stepping outside the platform, he weighs in.

Less Code, More Leaps, Some Caution

Evolution is taking care of a lot of issues discussed above. As Joysula captures: Features such as live code sharing, multi-cursor editing, reviewing and commenting and controlling roles and access address the challenge of developer control. “Further, it is practical to have an à la carte of approaches, tools and platforms to meet the diverse application requirements.”

Not all issues are beyond resolution though.

As Weaser shares, developer control, for instance, is preserved through DevOps tools and extensible code extensions, enabling proficient developers to customise solutions while still using visual development for efficiency. “Flexibility is also less of a constraint on advanced low-code platforms, which support custom logic and deep integrations to address a variety of use cases. Security worries, often one of the top concerns, are also addressed through comprehensive measures like encryption, compliance certifications, and automated vulnerability scanning – all of which ensure data protection and regulatory adherence.”

Many modern low-code and no-code platforms have significantly improved in the areas discussed above, Wong also contends. “However, it’s crucial for organisations to carefully evaluate platforms to ensure they meet their specific needs regarding scalability, flexibility, and security.”

For the rebuild stress, one can consider internal tools and dashboards and greenfield applications that require minimal or no rebuild.

The freedom to build without IT involvement can lead to duplicated systems, hidden tools, and technical debt. That’s why training, review processes, and governance are essential for scaling safely and sustainably. Each type works best in different contexts, recommends Chalise—low-code for enterprise-grade applications, no-code for fast experimentation, and AI tools for accelerating development.

And the biggest concern is – humans forgetting how to drive a car. Are we ready for a world where “not-learning-coding” will be ok? AI coding tools enhance productivity but cannot replace foundational programming knowledge, concludes Chalise. “Relying too heavily on them may hinder learning, especially for junior developers.”

It’s a strange world that we are heading to. A world where early buyers of autonomous cars co-exist with those who still love the good old skill of horse-riding. Sometimes, they are even the same set of people. Only on different tracks and on different parts of their weeks. And until we see pilots and astronauts playing Solitaire – all the time – while flying planes and spaceships; and unless we see enterprises running apps with huge scale and stakes without an IT eye – nothing is ‘TLTR’. Only ‘IYKYK’.

(If You Know, You Know!)

pratimah@cybermedia.co.in