The participation of LinkedIn Co/AI in the “VIBE COMER Frontue Deleper” list, is a sign that software development is changing rapidly. This emerging practice “Costume Vepi” The development of writing a sentence turns into talks with artificial intelligence, and occurs faster than many people.
The upscarcated programmer climbing
In February 2025, Andrej Karpathy, the former AI’s first manager in Tesla and the famous computer science, was coined by describing a different approach to the creation of software. Instead of writing a sentence systematically, developers now describe what they want as natural and allow the generation of specialized artificial intelligence models.
“I just see things, say things, run things, copy, and mostly work.”
This informal description of Karpathy carries a deep shift that already changes who can create programs and how to build it.
From writing to the hadith: How does the coding work
The beauty of coding lies in its simplicity. Instead of writing the code line line, developers:
- Describe what they want In a clear language, as explained by a colleague
- See what artificial intelligence generatesOften multiple implementation options
- Submit notes On what is doing and what is not.
- Reception through conversation So that the symbol does exactly what is required
“It is like the presence of a young developer who never tired and read every rule of code written at all,” One startup told me. “You direct them in a clear language instead of building a sentence.”
The ecosystem of tools is rapidly evolved to support this workflow, with GitHub Copilot, an indicator, re -V0, Claude, kokrom, beloved, Bolt and many others that allow this approach to conversation in development.
Why are companies betting on coding the atmosphere?
The numbers tell a convincing story. A quarter of the products integrated in the winter of 2025 in the winter of Y Combinator with the code bases that are created 95 % of artificial intelligence. The effect goes beyond individual productivity. Garry Tan, the CEO of Yombinator recently, told Business Insider that “VIBE coding” allows startups to remain smaller, with 10 engineers now able to do the work that previously required a team from 50 to 100 people.
The first adoption reports significant gains in productivity, while completing some teams in weeks that may take a traditional traditional.
What makes the wonderful programmer?
Success in this new model requires a different set of skills on traditional development. The most effective practitioners bring together:
- Clear contact – The ability to clarify technical requirements accurately
- Learn about patterns – Discover quickly when creating a code that needs to be improved
- Systems thinking Understand how ingredients should interact at a high level
- UX intuition Give priority to how the final product feels users
I found in my own experience as a great engineer who embraced this approach.
“Technical knowledge is still important, but it is more important to know the shape of a good code than you are able to write each line yourself.”
Where the coding shines (and where it does not happen)
Not every project is suitable for coding. The approach works beautifully for:
- Fast initial models Build MVPS and quickly test concepts
- User interfaces – Create a modern, responsive forehead
- Standard – Implement common features such as ante or payment processing
- Solo projects Enabling individual developers to build what usually requires the difference
But the challenges appear when dealing with:
- Complex algorithms – Where it concerns sports accuracy.
- Important systems Health care or financial systems, where it can be severe consequences and a complete understanding of each necessary line
- Security sensitive ingredients Approach or data protection systems, as AI may not be inadvertently advanced weaknesses.
- The integration of the old system – Working with the old code is poorly documented, as artificial intelligence lacks the context about unrelated behaviors or solutions
I recently spoke with a cloud architect participating in a realistic experience with full coding: I was helping a customer to create a default device scale on Azure using the infrastructure code created by artificial intelligence. Everything looked perfect and successfully spread it, but artificial intelligence has completely missed the formation of the decisive safety certificate. The censorship was not arrested until the security team performed standard checks – and that is the point they already paid to production. It could not have been clearly clear to the experienced developer that was completely missed in the code created by artificial intelligence.
This highlights the decisive challenge: Amnesty International can create a business code that passes the initial test, but it still contains safety issues or basic compliance that only field experts recognize. Suitable review protocols are necessary.
One of the senior developers told me during a recent meeting: “We use the coding of me for about 60 % of our blade base.” “We usually use artificial intelligence to create the first code and then work on it. For the most important parts, we still rely on traditional development. It comes to using the correct approach for each component.“
What does this mean for your professional life
If you are a developer wondering how to adapt to this shift, your approach should differ based on your experience:
For junior developers:
- Start using GitHub Copilot (free for Everone) to enhance your learning process
- Focus on understanding the code that AI generates instead of using it only
- Build side projects using extremist coding to develop your rapid engineering skills
- Jo.
For developers of the average level:
- Merging artificial intelligence tools into your current workflow without giving up traditional skills
- Focus on developing experience in evaluating code created from artificial intelligence
- Building a portfolio showing both traditional coding capabilities and artificial intelligence cooperation
- Try using artificial intelligence for the coding parts that you find boring and harassed for creativity
For senior developers and technical technology:
- Developing strategies to integrate coding safely into the workflow
- Create a specially designed review of the symbol created from artificial intelligence
- Focusing on architecture skills and system design that is still artificial intelligence struggle with
- Training your team on effective rapid engineering and artificial intelligence cooperation
Everyone’s key balance. One of the developers, who recently obtained a new job in a start -up company, also told me:
“Our team still writes traditional software instructions when it is logical, but we can now build 5x faster by knowing when and how to take advantage of artificial intelligence.”
The road forward
While we look forward to the future, it is clear that coding in the Libyan is not just a transient direction. Companies like CO/AI are betting on their technical strategy, and productivity gains are very important to ignore them.
The question is not whether artificial intelligence will transform programming – it is already happening. The real question is –
How we will adapt our skills, teams and development processes with this new reality as the writing code has become a conversation.
Developers who are mastering this mix of conversation and symbol will not adapt to the future – they will define it. As the line between the human and the AI continues to blur, those who can speak two languages will build the next generation of technology that forms our world.
What is your experience with coding Vepby? Have you tried any of the aforementioned tools? Share your thoughts in the comments below!
Reference
Techcrunch. (2025). “A quarter of the startups in the current YC group have a symbol rules that are almost completely created.”
Business Insider. (March 17, 2023). “The coding is 10 engineers to do the work of a team between 50 and 100,” says CEO of Silicon Valley Invicator.
If you find this article value, check my previous virus’ article on NLC, which got readings ~ 13K and my last article about Trending Form context protocol (MCP). The scene of artificial intelligence develops rapidly – knowing that it remains at the forefront.
About the author: I am Jay Thakur, the chief software engineer in Microsoft, to explore the transformational capabilities of artificial intelligence agents. With more than 8 years of experience in building AI solutions and expanding its range in Amazon, Accenture Labs, and now Microsoft, along with my studies at Stanford GSB, bring a unique perspective of the intersection of technology and business. I am devoted to making artificial intelligence accessible to everyone – from beginners to experts – focusing on building influential products. As an ambitious speaker and consultant, I share visions around artificial intelligence agents, Genai, LLMS, SMLS, AI responsible, and advanced landscape. Contact me on LinkedIn and follow me on X.
