Vibe Coding: The Easiest On-Ramp & Upskilling for Coding Today

I’ve spent the past decade working in venture capital meeting with hundreds of very technical teams, and I’ve always been a bit self-conscious about not knowing how to code.

A few years ago, I tried my hand at Python through online courses—dabbling in automation and data processing to speed up work and for small hobby projects.

Then, last year, I discovered Vibe Coding—using LLMs like ChatGPT to write and refine code. It’s been heralded as both the future of coding (by the world-famous accelerator program Y Combinator) and the end of coding (with the challenge of #vibedebugging and code bases becoming incomprehensible).

I will argue below why I think it’s the best on-ramp to tech the world has ever seen.

What Have I Built?

For work:

• A system to send personalized startup recommendations to hundreds of investors in our virtual events.
• A tool to create 100+ Google alerts automatically for our massive portfolio (using GUI automation and image recognition - as Google didn't have an API and blocked browser automation).
• Plenty of one-time apps to scrape or process datasets and automate tedious workflows.

For play:

• A flappy bird game where the bird eats items with different glycemic index and has to avoid diabetes.
• A telegram bot to track a crypto asset hourly.
• An app to track baby feeding time and mood.

But that’s not the point. The real value of Vibe Coding isn’t in what I’ve built—it’s in how easy it is to build, by anyone. Getting results in minutes instead of hours, days or weeks (often after months or years of learning to code).

Coding Has Always Gotten Easier—Now It’s for Everyone

Programming has never stood still. It started with machine code, then moved to compiled languages, then high-level programming, then AI copilots. Now, Vibe Coding is the next evolution—turning coding into a conversation rather than a painstaking technical and syntactical exercise.

This isn’t just a curiosity for software engineers. It’s a practical superpower for non-technical professionals. Instead of hiring a developer or subscribing to yet another SaaS tool, anyone can now use AI to generate working code, tweak it, and deploy it—all with minimal prior experience.

More People Should Learn to Code, Not Fewer

Maybe there is no point learning to code if AI does it better? That’s a dangerous misconception. Andrew Ng , one of the most respected AI entrepreneurs and educators, made this point clear:

“Some people today are discouraging others from learning programming on the grounds AI will automate it. This advice will be seen as some of the worst career advice ever given. As coding becomes easier, more people should code, not fewer!”

Yann LeCun , VP and Chief AI Scientist at Meta, and a pioneer of deep learning, also emphasized the importance of understanding the foundations:

“We might have super-intelligent AI assistants in a while, but we’ll be their boss. We will need to understand the foundations in order to lead them.”

Sure, most of us won’t need to build cutting-edge AI models from scratch. But we can learn to wield this new “electricity moment” to our advantage.

Vibe Coding as a Teacher

While some will use AI-generated code blindly, I found that Vibe Coding is also an excellent teacher. Because AI handles the grunt work, I could focus on how things actually work. Over time, I picked up:

• How to use all kinds of libraries and APIs to build powerful programs.
• How to organize, optimize, and secure code for sharing with others.
• How to use multiprocessing, for instance to simulate 10 million games of Werewolf in 30 seconds to finally know who wins. Spoiler: Werewolves (most of the time).
• How to recreate “Hot Dog, Not Hot Dog” from HBO’s Silicon Valley in just five minutes.
• How to explore advanced topics like computer vision and machine learning—enough to get my computer to play my Flappy Bird clone in its own “Hyperbolic Time Chamber” – about 200 times faster than human time – or research how to translate baby cries.

Along the way, I also picked up virtual environments, environment variables, and version control (though I admit I still rarely use Git properly). I know just enough to be dangerous. And now, so can you.

Further, it now allows me to venture beyond my Python comfort zone, with attempts at creating mobile apps, that LLMs are just as familiar with!

Vibe Coding Is the Easiest On-Ramp and Upskilling Path for Coding—And It Takes Just One Hour to Prove It

The best part? You don’t need months of training to see results. After learning on my own, I introduced Vibe Coding to friends and colleagues and eventually built a 1-hour crash course to share it more broadly. The goal isn’t to turn people into software engineers—it’s to build up their “talent stack,” give them more control over technology, and help them get things done.

I expected non-coders—entrepreneurs, creatives, analysts, and operators—to be the main audience. For those who had never coded before, the moment they saw their first AI-assisted program run was a breakthrough (“I never wrote a line of code—truly satisfying when you see your program finished and working!”); another student found that it unlocked creativity ( “I keep getting distracted because I keep getting ideas from this. I built an app for managing our fencing club members in between watching bits.”).

But I also realized that non-AI coders and ex-coders-turned-PMs were benefiting too. One of my students, a product manager, put it best: “I haven’t been coding for a few years now. This is frankly astonishing as a ‘look at what you can do (and how easily)!'”

Because in the end, coding has always been about solving problems. Now, with AI, solving problems is easier than ever. And once you start Vibe Coding, you’ll never look at technology the same way again.

ABOUT THE AUTHOR

The author is a senior partner with over 10 years of experience at a global venture capital firm. While he graduated as a mechanical engineer, he only learned to code online as a hobby recently and has since then embraced vibe coding. He has written previously about venture capital, startup ecosystems and been quoted in publications such as TechCrunch, WIRED, Bloomberg and more.