AI software engineers and the fear of disappearing job titles
AI software engineers and AI coding tools describe systems that can generate, modify, and maintain code for applications, shifting software engineering careers toward higher-level design, product thinking, and collaboration with automated coding agents rather than manual implementation of every feature. That shift is driving new worries about AI job displacement and the future of the developer job market. Boris Cherny, creator and head of Claude Code, argues that for some categories of work, coding is already “solved.” He has not written a line of code himself in more than six months and expects the title “software engineer” to begin fading, replaced by broader labels such as “builder” as designers, product managers, and leaders ship production code with AI support. The paradox is that the same tools raising fears of lost roles may also enable more people to participate in software creation than ever before.
From AI job displacement to a ‘golden age’ for founders
The rise of AI coding tools sits at the center of a split narrative about software engineering careers. On one side is AI job displacement: as tools write more code, traditional roles shrink or morph into oversight and orchestration. On the other side, these same tools lower the barrier to founding a company. Cherny’s advice to 22-year-old computer science graduates captures that tension. He notes that entry-level roles still exist, but urges anyone with an entrepreneurial streak to “go start a startup,” arguing that there has never been a better time to build. AI agents can now handle large parts of implementation and scaling, letting small teams move with speed that once required far larger engineering headcounts. In that sense, the developer job market is not disappearing; it is tilting toward ownership and product creation.
How AI coding tools are transforming the developer job market
Evidence from early-stage founders hints at how thoroughly AI coding tools are changing work. Cherny says that when he spoke to the latest Y Combinator cohort, about half the founders in the room reported that Claude Code writes 100% of their code. At the same time, only a single person out of a few hundred said they avoid AI-generated code entirely. Everyone else was somewhere between those extremes, suggesting that AI software engineers—in the form of coding agents—are already standard teammates. That adoption reshapes the developer job market: the most cutting-edge startups now treat code as a largely automated asset, while humans focus on architecture, product strategy, and user understanding. Coding is “getting solved” for a larger share of routine tasks, but oversight, debugging, and integration of AI-created systems remain human responsibilities, supporting a shift in roles instead of simple elimination.
New roles, new skills: why software engineering careers are not over
Despite references to the “end” of software engineers, Cherny is clear that the work of building software is not vanishing. He expects that people who write code or direct agents to write code will expand dramatically in number. His prediction is stark: there will be 100 times more such people in the future than today. That implies a broader, more inclusive developer ecosystem in which non-traditional profiles participate. Sam Altman has made a similar point about what matters on founding teams, saying that people who understand users well but cannot code are now attractive candidates for funding when they pair with AI coding tools. New roles—agent orchestrators, AI product leads, and domain experts who guide AI systems—will grow alongside classic engineering specialties. The job description is changing, but for those willing to learn AI tooling and develop product sense, opportunities are multiplying rather than shrinking.