Back to blogs

Vibe Coding: My Journey and Thoughts about vibe-coding

Exploring the concept of vibe coding and its impact on my development journey

3 min read
By Benjo Quilario

Vibe coding is a term that has been circulating in the software development community, and it resonates deeply with my own journey as a developer.

What is Vibe Coding?

Computer scientist Andrej Karpathy, a co-founder of OpenAI and former AI leader at Tesla, introduced the term "vibe coding" in February 2025. The concept refers to a coding approach that leverages large language models (LLMs), allowing programmers to generate working code by providing natural language descriptions instead of manually writing it.

My Journey with Vibe Coding

I've encountered the term "vibe coding" frequently on Twitter, YouTube, and other social media platforms. However, it wasn't until I started using it in my own projects that I truly understood its significance. Vibe coding has transformed the way I approach software development—experiencing it firsthand has been both exciting and a little intimidating.

I decided to build a chat application using vibe coding.

After setting up my environment, configuring Prisma, and organizing the folder structure, I prompted an AI agent:

Help me build a chat application using Prisma and Pusher based on my current directory structure. Please review my prisma/schema.prisma file to ensure it is well-designed. The application should feature appealing animations and automatically load more messages when the user reaches the end of the current chat viewport. Additionally, please ensure that all server actions are saved in the database folder.

It felt like magic. A project that would normally take months to build suddenly seemed possible in just days. The AI agent understood my requirements and generated the necessary code snippets, including the Prisma schema, Pusher integration, and even the frontend components with animations. It was as if I were collaborating with a highly skilled partner who could seamlessly translate my ideas into functional code.

I was both amazed and a bit apprehensive. If AI can do this, what will happen to us as developers? Will we still have jobs? Will we still be needed in the software development industry? As students, will we still be able to find opportunities? These questions lingered in my mind as I continued to explore vibe coding.

Of course, the AI-generated code wasn't perfect—there were still bugs and issues that needed to be addressed. You still need to review the code, understand it, and make necessary adjustments.

The Future of Vibe Coding

AI has advanced rapidly, and its progress is unstoppable. Rather than fearing these changes, we should focus on leveraging AI as a tool, ensuring that we use it to our advantage—not the other way around. Vibe coding marks a significant shift in how we approach software development. It’s no longer just about writing code; it’s about collaborating with AI to build better, more efficient solutions. As we embrace this new paradigm, it’s crucial to maintain our core skills and a strong understanding of fundamental software development principles.

Thoughts

Vibe coding is a powerful tool that can enhance our productivity and creativity as developers. However, it should not replace our fundamental understanding of programming concepts and best practices. We must continue to learn, adapt, and grow in this ever-evolving field.

Vibe coding isn't about blindly trusting AI; it's about leveraging its capabilities to enhance our own skills and creativity.

Do not rely on vibe coding alone. As your codebases grow, it will become much harder to fix issues—especially if you encounter problems that vibe coding and prompting can't solve. Fixing them can become a nightmare. You need to understand the code and the underlying principles to effectively debug and maintain your projects.

Related Posts