manifesto - dev of the future.

every time we open LinkedIn, Reddit, or X/Twitter, we see posts from experienced tech professionals struggling immensely to find a new job. the situation is even more brutal for beginners: young adults who just entered college, looking for their first opportunity, with no idea how to compete with the sheer speed at which AI and the market are evolving.

the truth is, there’s no sugarcoated answer. adapting hurts, but it’s the only path to survival.

inspired by discussions around the concept of the Full-breadth Developer, I’m writing this manifesto on what I see as the profile of the tech professional of the future.

for the general public, the revolution of what these new technologies will actually accomplish hasn't even truly begun. but if you work or want to work in tech, you’re already behind. the best time to master these tools was yesterday. the second best time is today.

if you’re reading this in 2026, here is the reality we’re facing:

a few months ago, I wrote that any statement about the future of the market with AI was nothing more than speculation, because no one really knew how far the tech would go. that premise remains true in the sense that the evolutionary curve is unpredictable, but one thing has become an absolute fact: there is no longer any room for skepticism regarding AI's role in software engineering.

anyone following the industry daily has seen the public adoption of AI in global-scale projects used by billions of people. the list is endless, but just look at Linus Torvalds, Markus "Notch" (creator of Minecraft), Kent Beck (co-author of the Agile Manifesto)... from software architecture pioneers to technical founders, pick your favorite computer scientist.

what do they all have in common? they leverage AI in the systems they build.

my single certainty about the market is this: AI for developers is no longer optional. you either adopt it, or you’re out. the option to work in this field without knowing how to orchestrate and generate solid code using these tools no longer exists—and never will again.

but where, exactly, does that leave us?

what has actually changed?

I. the fall of the dual-track system

for a long time, the market—both academia and companies—trained us to believe that software development had to be rigidly split into two distinct roles:

• designers / product managers (the visionaries who conceive the idea)
• software engineers (the builders who type the code)

they created Y-shaped careers and dual ladders: "tech folks sit over there, product folks sit over here." suggesting that a single person could drive both product design and technical execution felt like an operational absurdity.

AI has shattered that wall.

the barrier between "idea" and "implementation" has dissolved. modern generative code tools have crossed a capability threshold where coding execution speed is no longer the primary bottleneck. the true bottleneck has shifted back to what it should have always been: critical thinking, good taste, business sense, and strong judgment.

if yesterday's top developer was a virtuoso violinist performing solo, today's top developer is the conductor leading an entire orchestra of intelligent agents.

II. the illusion of isolated roles

when we analyze the impact of AI through the lens of traditional roles, it becomes clear why the old structure is collapsing:

• the traditional junior developer:

lacks enough context and mileage to evaluate the quality of AI-generated code. faced with AI hallucinations, beginners risk internalizing flawed concepts. ironically, the spike in raw volume of AI-generated "lines of code" has been slowing entire teams down, forcing them to spend days debugging and reviewing what has come to be called "AI slop" (low-quality code pushed without discernment).

• the purely technical senior developer:

can prompt AIs to generate acceptable code extremely fast. however, if they limit themselves to "coding for the sake of coding" and wait around for handed-down requirements to go through the motions, they become replaceable. competing on pure execution against fleets of relentless, cheap autonomous agents is a losing battle.

• the non-technical product manager:

can hack together quick prototypes and working apps using no-code tools or LLM assistants. however, without a deep grasp of underlying system architecture, any system bigger than a toy project eventually collapses under its own weight. the passive compliance of AIs combined with naive product optimism leads to endless cycles of rework and frustration.

• the rise of the dev of the future

this is where the winning profile emerges: the Full-breadth Developer (or as I like to put it simply, the "dev of the future").

III. who is the dev of the future?

this professional isn't defined by years on a resume, but by mindset. they pair rock-solid technical skills with sharp product sense and business acumen.

if you put a dev of the future to work, they can deliver in two days a backlog that would typically take weeks of alignment between a "product manager" and a "coder." internal communication costs simply vanish when product vision and technical execution live inside the same mind.

the key traits defining this professional in the new landscape are:

• outcome orientation (ROI): they love technology, but they love solving real problems and driving business or operational value even more. they know exactly where code touches revenue or customer experience.
• technical rigor and discernment: they don't blindly accept the AI's first response. they act as an unyielding editor, auditing code, enforcing architectural standards, and understanding precisely why something works.
• initiative and proactivity: they don't sit around waiting for a polished PRD (product requirements document). they spot product gaps, experiment with solutions, and actively pitch improvements.

IV. a diagnosis for anyone looking to survive

if you want to stay relevant and well-compensated in this industry, the golden rule has shifted: position yourself as close to value creation as possible.

the longer the sentence required to explain how your day-to-day job makes or saves money for the company, the further you are from real value—and the more vulnerable you become.

V. for those just starting out:

the good news is that traditional coding education was always too limited to shape well-rounded professionals anyway.

when used with discipline, generative AI tools offer the most powerful learning environment ever built.

don't use AI as a crutch to copy and paste answers;

use it as an unrelenting personal tutor to understand the why behind architectural choices, challenge its outputs, and accelerate your ability to build real-world products.

VI. for industry veterans:

drop the defensive attitude of hiding behind a job description to avoid dealing with product or business decisions. master the ecosystem of advanced AI agents and developer tools, take ownership of the final impact of your software, and prove that your critical judgment is the exact engine that turns generative tech into true business value.

conclusion

the era of the programmer who receives a locked scope of work and mechanically turns it into code without understanding the business context is over.

the tech industry isn't dying; it's undergoing a long-overdue filter. the winners of this new era will be those willing to play every position on the field: shaping the vision, auditing the execution, and delivering software that actually drives results.

the division between Product and Tech is officially dead.

welcome to the era of the Full-breadth Developer.