Software Loses Value

Currently, making software is so cheap that its market price is at rock bottom. It’s a wave that takes time to reach some sectors, but it eventually arrives. As I try to organize my thoughts on this disruptive landscape, I find a clear analogy: the value of aluminum. There was a time when this metal was highly prized until its abundance was discovered and processing methods were optimized. As it became hyper-abundant, its price plummeted. The exact same thing is happening to software.

The Paradox of Ideas and Technology

Today, anyone with a potentially millionaire idea has the tools available to build their own applications. However, no one has become a millionaire just because of that yet.

The answer lies in the hierarchy of the process. Beyond the idea is the business, and beyond the business, technology. The ideation and basic technology phases are practically resolved. The real challenge I face when building projects lies firmly in the design and sustainability of the business model.

Community as a Business Driver

Building a business is currently the bottleneck. After years of structuring solutions, I have found that the most efficient strategy in today’s environment is to create a community. This approach allows for controlling the entire commercial flow:

• Attraction: You create an audience large enough that values the content and technical vision you share.

• Launch: You introduce a product designed exactly to solve the needs of that audience.

• Scalability: The larger the follower volume, the greater the number of buyers, increasing revenue even without changing the conversion rate.

The strategic approach has mutated. Before, it was enough to launch a technically excellent product; today, that product needs a community to support it. And the only way to attract that community is by generating content that genuinely responds to their interests.

The Empty Terminal Crisis

I have dedicated many years, hours, and effort to creating software. Today, when it’s cheaper and simpler than ever, I sometimes feel stuck. I feel that conventional projects no longer deserve to be developed from scratch by such powerful tools. I often find myself thinking, in front of the open terminal with Claude, what I should build next.

Everything made more sense to me when things presented friction. Implementing a simple project could take days or weeks of deep research. It forced me to master multiple layers of abstraction. Today, I doubt it’s necessary to implement utilities like Tailwind manually if an AI can instantly generate the CSS structure.

Agents as the New Compilers

What remains for us in the face of this uncertainty is to sell better, tackle problems of greater architectural complexity, and dare to execute what others avoid. Frameworks will soon cease to be tools for humans and will become infrastructures for autonomous agents.

I understand agents as the new compilers. Instead of translating from a structured programming language to machine code, they will translate natural language into executable code. It’s just the next step in the evolution of our tools. The difference is that giving precise specifications in natural language is not the same as everyday speech. Therefore, Requirements Engineering regains its critical value: it will be the discipline responsible for translating ambiguous needs into exact instructions.

Competitive Moats Beyond Code

I assume that software is no longer the business; it is strictly the operational means for the business to function. If a SaaS is easily replicable, any intelligent engineer can generate an identical or better-adapted version. In light of this, I identify three key competitive moats that cannot be replicated simply by automating code:

• Proprietary Data: If you possess data sets that no one else has and are difficult to extract, you have an immediate advantage.

• Technology and Physical Integration: Utilizing resources that operate outside the pure software ecosystem, such as specialized hardware or engineering materials, imposes a very high replication barrier.

• Deep Business Knowledge: Expert and intimate understanding of the sector allows for designing solutions so aligned with operational reality that their adoption is natural and almost mandatory.