At Being Guided we are bringing to market a new AI innovations that replace an earlier generation of Software-as-a-Service (SaaS) apps. So, firstly, what is Vibe Coding¹, secondly, who's it for, and thirdly, why should you care? Here are the answers:

1. Vibe Coding is a terms first used by industry analyst Andrej Karparthy¹ to describe how you can generate software code by using AI to describe apps using natural language, where the term 'vibe' comes from "fully giving into the vibes, and forget that the code even exists".
2. Vibe Coding is for everyone. Even our largest customer, the UK National Health Service (NHS) is embracing next generation Software-as-a-Service (SaaS) apps built with AI-powered Vibe Coding. Elimating a £604million deficit¹¹ at the NHS has some influence here!
3. You can now avoid or replace expensive SaaS licences by engaging with AI-powered technology. We become your Vibe Coding Partner, empowering you to move beyond expensive first generation SaaS apps, using best-in-class AI technology.

Does history repeats itself?

Having worked with Salesforce technology for the past 20 years I remember the early stages of this Software-as-a-Service (SaaS) pioneer: it was simply faster, simpler and cheaper than what went before: namely, Siebel on-premise Customer Relationship Management (CRM) systems.

Fast-forward 20 years and just as Siebel was displaced by Salesforce as the global leader in CRM systems, today, we see AI-powered VIbe Coding as the disrupter of Salesforce (and other first generation CRM apps) for the same reasons: it is simply faster, simpler and cheaper.

No-Code. No problem.

Salesforce and first generation CRM SaaS apps are one example of where Vibe Coding with an AI-powered CRM Pro Framework can lead to creating and replacing many SaaS apps. To quote leading industry analyst at Forrester, John Bratincevic¹⁰:

“Nobody wants to admit that their own death is coming soon. Low-code has been swinging the pendulum away from off-the-shelf applications and toward custom development for years. There are good reasons for this. When practical, fit-to-purpose software is best. And the lower cost, risk, and lead time of low-code development - coupled with an expanded developer pool, easier integration, management of apps on a common platform, leveraged licensing, etc. - makes it much harder to justify off-the-shelf software licenses and vendor sprawl. AI-powered enterprises will 'build' software instead of 'buy' it — and many applications in enterprise portfolios will consolidate onto low-code AppGen platforms."

Emerging role. Vibe Coder.

For AppGen now read Vibe Coding. For low-code now read the strapline for Vibe Coding: 'No code. just words'. As illustrated above, I see AI-powered CRM apps as replacing Salesforce and other first generation Software-as-a-Service (SaaS) apps.

Vibe Coding for creating CRM and other SaaS apps is an AI-driven, No-Code First approach to software development. This is where users can describe their desired CRM or other SaaS app in plain, natural language prompts or instructions.

From natural language inputs, AI automatically instantly generates a fully functioning SaaS app, complete with frontend, backend and supporting infrastructure, without requiring manual coding or technical expertise.

This process leverages advanced Large Language Models (LLMs) with integrated development tools to translate high-level ideas into executable code. This is shifting the Vibe Coder’s role from traditional programming to guiding and refining AI-generated outputs.

AI-powered Vibe Coding should serve as an augmentation tool for Software Developers rather than as their replacement, but fewer required, nevertheless. The emergence of the Vibe Coder is going to creating a new environment for co-designing apps.

Think of digital innovation and software publishing now as more business analysis, less software development. A Vibe Coder is more like a business analyst or subject matter expert, than a tech architect or programmer.

A real world example.

Let's explore this further. In this example we are Vibe Coding a CRM app. Of course, this can extend to other apps on a common business logic: Project Management; Talent Management; Professional Services Automation, and so forth. 'CRM Pro' then becomes 'CRM Plus'.

The first step here is engaging in a Mutual Value Discovery to determine where Vibe Coding can generate next generation SaaS app at a fraction of the cost of existing technologies, yet must be result in a solid user experience and secure, scalable IT.

At Being Guided we are now helping enterprises adopt Vibe Coding for next generation CRM apps, where non-programmers, or Vibe Coders, collaborate with Software Developers under the guidance of an AI Agent. It's a new kind of 'Pair Programming'.

The AI technology includes:

• Database Setup
• AI Large Language Models
• Design System
• Email System
• Authentication
• Analytics
• Storage

AI-powered Vibe Coding is capable of generating code across the entire software development stack. This starts with prototyping with an inherent ability to quickly translate natural language prompts into fully functional code (Chen et al., 2021³).

Vibe Coded apps generated by AI must meet key questions around potential issues: security vulnerabilities, or performance bottlenecks - all critical in production-class applications (Dahlin, 2021⁴). So our Discovery Engagements at Being Guided are truly minful of these realities.

At Being Guided we are creating an App Framework called 'CRM Pro'. Here we started with a simple Data Model of typical Entities (Objects) you would find in a CRM app: Leads; Contacts; Accounts; and, Opportunities.

Reality check.

Production-class CRM apps must meet stringent requirements for security, scalability, and maintainability. These apps often handle sensitive data, integrate with complex systems, and serve large user bases - all of which demand a level of precision and reliability, including:

• Security Risks: AI-generated code must avoid introducing vulnerabilities, such as SQL injection or cross-site scripting, due to a lack of context (Perry et al., 2022⁷).
• Scalability Challenges: AI tools optimised for performance and anticipating need for horizontal scaling, avoiding bottlenecks as user demand grows (Bass et al., 2019²).
• Business Logic Complexity: AI fully grasping the business rules or compliance requirements, which are often context-specific, require human judgment (Holstein et al., 2019⁵).
• Maintenance and Debugging: AI-generated code must be easy to maintain, including sufficient modularity and docs that human developers prioritise (Lwakatare et al., 2020⁶).

Working in pairs.

Our approach to Vibe Coding is Pair Programming: where a non-programmer (known as Vibe Coder) collaborates with one or more Software Developers to co-create a CRM app. Guided by AI, the Vibe Coder contributes domain expertise and high-level requirements, whilst the Software Developer ensures technical rigour.

Pair Programming leverages the strengths of both parties: the Vibe Coder’s understanding of the problem space and the Software Developer’s ability to implement scalable, secure solutions (Holstein et al., 2019⁵).

Over time, the ratio of Vibe Coders to Software Developers will change. The productivity gains from AI-powered Vibe Coding will become more measurable. This will mean more Vibe Coders, less Software Developers.

My conclusion.

Vibe Coding takes advantage of best-in-class Large Language Models (Anthropic Claude Sionnet 4, Google Gemini 2.5 Pro, ChatGPT4, etc.). Vibe Coders can run natural langage prompts across multiple Large Language Models.

Design Thinking⁷ and Value Engineering⁸ underpin Vibe Coding in our work at Being Guided. Design Thinking maximises stakeholder engagement by generating empathy and trust. Value Engineering builds robust business cases using ROI Models to defend value and quantify costs.

Consider a scenario where a knowledge-intensive services company needs a custom CRM app to manage client portfolios. A Vibe Coder describes the app’s requirements in natural language, and AI generates the code.

At Being Guided our approach is to use collaborative methods, where Vibe Coders and Software Developers work together, guided by a Large Language Model (LLM) powering the AI.

A paired Software Developer then refines this platform code, ensuring it meets best practices, security standards and integrates with existing business logic. It's an iterative, safe journey from first generation CRM, such as Salesforce, to next generation Vibe Coded SaaS apps.

Lets remember the motivations for change: faster, simpler and cheaper.

Faster. Vibe Coding is inherently many times faster in the design, development and delivery phases of Web apps when compared to customising Salesforce or other earlier generation Software-as-a-Service (SaaS) apps. It's days and hours versus months and weeks.

Simpler. At Being Guided we apply Fierce Reduction: the practice of simplifying all business processes, tasks and information systems by removing redundant or non-essential elements before considering Vibe Coding for next generation app development with AI.

Cheaper. With Salesforce Sales Cloud the annual subscription fees for licences is £120 per user per month for the Enterprise Edition. Added to this are professional services fees, often many times the annual subscription fees. With AI-powered CRM Pro apps, tailored to you exact needs, the costs are a tiny fraction of Salesforce licences. The ROI Model is truly compelling.

Vibe Coding with Rocket. Natural lanaguage prompts on left. CRM Pro app on right.

Next?

Let's Meet to explore your Vibe Coding - one hour at a time.

References

1. Karpathy, A. [@karpathy]. (2025, February 2). There’s a new kind of coding I call “vibe coding”, where you fully give in to vibes, embrace exponentials, and forget that code even exists. X. https://x.com/karpathy/status/1753472166197080428
2. Bass, L., Clements, P., & Kazman, R. (2019). Software architecture in practice (4th ed.). Addison-Wesley.
   https://www.oreilly.com/library/view/software-architecture-in/9780136885979/
3. Chen, M., Tworek, J., Jun, H., Yuan, Q., Pinto, H. P. D. O., Kaplan, J., ... & Zaremba, W. (2021). Evaluating large language models trained on code. arXiv preprint arXiv:2107.03374.
   https://arxiv.org/abs/2107.03374
4. Dahlin, M. (2021). The role of AI in software engineering: A review. Journal of Systems and Software, 174, 110887.
   https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3633525
5. Holstein, K., Wortman Vaughan, J., Daumé III, H., Dudík, M., & Wallach, H. (2019). Improving fairness in machine learning systems: What do industry practitioners need? Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems, 1-16.
   https://arxiv.org/abs/1812.05239
6. Lwakatare, L. E., Raj, A., Bosch, J., Olsson, H. H., & Crnkovic, I. (2020). A taxonomy of software engineering challenges for machine learning systems: An empirical investigation. Journal of Systems and Software, 162, 110496.
   https://colab.ws/articles/10.1007%2F978-3-030-19034-7_14
7. Perry, D. E., Porter, A. A., & Votta, L. G. (2022). Empirical studies of software engineering: A roadmap. ACM Computing Surveys, 54(3), 1-36.
   https://dl.acm.org/doi/10.1007/s10664-009-9121-0
8. The Hasso Plattner Institute of Design. (2004) Stanford d.school. https://dschool.stanford.edu/about
9. Miles, L.D. (1947). The Lawrence D. Miles Value Engineering Reference Center Collection.
   https://minds.wiscon.edu/handle/1793/301
10. Bratincevic, J. (2024, November 13). AppGen is an existential threat to the enterprise app business. Forrester.
    https://www.forrester.com/blogs/appgen-is-an-existential-threat-to-the-enterprise-app-business/
11. NHS England. (2025, March 27). Financial performance update. https://www.england.nhs.uk/long-read/financial-performance-update-5/