← Retour au blog
tech 28 August 2026

A Million Kakapos: The Future of Agentic Coding

The age of agentic coding is redefining open source. What happens when coding becomes 'free' with LLMs? Discover the potential impact on the software ecosystem.

Article inspired by the original source
A Million Kakapos ↗ blog.mempko.com

Introduction

The field of software development is undergoing a revolution with the emergence of agentic coding. Since the advent of large language models (LLMs) like GPT-3, coding has become more accessible than ever. We are at a point where writing code is almost 'free'. This raises the crucial question: what is the future of open source software when coding is no longer a scarce resource?

Agentic Coding: A New Paradigm

Agentic coding refers to the use of AI models capable of generating code autonomously. This not only changes how code is written but also who can write it. LLMs now enable small teams to produce complex software without the need for expert developers. According to a McKinsey study, the adoption of AI in software development could reduce costs by 20% to 30% by 2025.

Impact on Open Source

Historically, open source has been the engine of innovation by providing freely accessible code for all. However, if code becomes a commodity, value will shift to other aspects like innovation, security, and maintenance. Open source projects might see a decrease in direct human contributions but benefit from an increase in LLM-generated contributions.

An Example: GitHub Copilot

GitHub Copilot, powered by OpenAI, is a striking example of how AI can integrate into the development process. In 2023, GitHub reported that Copilot contributed to over 40% of code in some popular open source projects.

Risks and Opportunities

As highlighted in the article on kakapos, over-optimizing for a specific environment can make a system fragile to change. Agentic coding must be used wisely to avoid over-dependence on models that may not adapt to new technologies or standards.

Security and Ethics

With increased automation, ethical and security concerns become critical. Models must be responsibly trained to avoid bias and ensure the security of the code produced.

Conclusion

The future of software development is promising with agentic coding, but it requires adaptation and strategic thinking to leverage its benefits while minimizing risks. Open source, as a concept, will remain relevant but will evolve to adapt to this new reality.

Let's discuss your project in 15 minutes.

Agentic Coding Open Source LLMs Software Development AI
Deepthix newsletter · 100% AI · every Monday 8am

An AI agent reads tech for you.

Our AI agent scans ~200 sources per week and ships the best articles to your inbox Monday 8am. Free. One click to unsubscribe.

Visit the newsletter page →

Want to automate your operations?

Let's talk about your project in 15 minutes.

Book a call