Andrej Karpathy, a founding member of OpenAI and former head of AI at Tesla’s autopilot division, has publicly acknowledged feeling “behind” as a programmer despite his pioneering role in the AI revolution. In a candid X post on December 26, 2025, Karpathy revealed that the software engineering profession is being “dramatically refactored” as AI tools increasingly automate code generation.
Karpathy, who coined the term “vibe coding” in February 2025—a practice where developers prompt AI to generate code rather than writing it manually—expressed that he could be “10X more powerful” if he properly leveraged the AI tools that have emerged over the past year. The Collins Dictionary even named “vibe coding” its word of the year, highlighting the term’s cultural impact.
The transformation Karpathy describes stems from revolutionary AI-powered development tools like Cursor, Claude Code, and Codex, which have fundamentally altered how software is created. Business Insider characterized 2025 as “the year coding changed forever,” reflecting the seismic shift in the industry.
Boris Cherny, creator of Claude Code for Anthropic and a leading figure in AI-assisted development, echoed Karpathy’s sentiments in the comments. Cherny admitted feeling this way “most weeks” and noted that he sometimes approaches problems manually before realizing AI can solve them faster. Interestingly, Cherny suggested that new graduates and early-career coders may adapt better to this environment because they don’t carry preconceptions about AI’s limitations.
Karpathy compared the current situation to receiving a “powerful alien tool” without a manual, while a “magnitude 9 earthquake” rocks the programming profession. He used vivid analogies to describe the unpredictable nature of AI coding tools—sometimes they “shoot pellets” or “misfire,” but occasionally “a powerful beam of laser erupts and melts your problem.”
Cherny emphasized that adapting requires “significant mental work” as models improve every month or two, with AI becoming increasingly capable at coding and engineering tasks. This rapid evolution means even AI pioneers like Karpathy must constantly recalibrate their understanding of what’s possible, highlighting the unprecedented pace of change in software development.
Key Quotes
I’ve never felt this much behind as a programmer. The profession is being dramatically refactored as the bits contributed by the programmer are increasingly sparse and between.
Andrej Karpathy, OpenAI founding member and former Tesla AI head, shared this striking admission on X, revealing how even AI pioneers struggle to keep pace with the tools they helped create.
I have a sense that I could be 10X more powerful if I just properly string together what has become available over the last ~year. A failure to claim the boost feels decidedly like skill issue.
Karpathy acknowledges the massive productivity potential of current AI tools while candidly admitting his own struggle to fully leverage them, framing adaptation as a necessary skill challenge.
Everyone has to figure out how to hold it and operate it, while the resulting magnitude 9 earthquake is rocking the profession.
Karpathy uses earthquake imagery to describe the seismic disruption AI tools are causing in software engineering, comparing them to powerful alien technology distributed without instructions.
It takes significant mental work to re-adjust to what the model can do every month or two, as models continue to become better and better at coding and engineering.
Boris Cherny, creator of Claude Code at Anthropic, highlights the cognitive challenge of keeping pace with rapidly improving AI capabilities, suggesting even experts must constantly recalibrate their expectations.
Our Take
Karpathy’s admission is remarkable because it reveals a paradox at the heart of the AI revolution: the creators of transformative technology can themselves become disrupted by its rapid evolution. His struggle represents a broader challenge facing the entire software industry—the need to fundamentally reimagine programming as AI-orchestration rather than manual code-writing.
The “vibe coding” phenomenon he pioneered has evolved so quickly that even he feels left behind, suggesting we’re in an unprecedented period of technological acceleration. The fact that newcomers may adapt better than veterans inverts traditional knowledge hierarchies and could democratize software development in unexpected ways. However, Karpathy’s weapon analogy—sometimes shooting pellets, sometimes firing lasers—captures the current reality: AI coding tools remain unpredictable and require sophisticated judgment to deploy effectively. The real skill may be knowing when to trust AI and when to intervene manually.
Why This Matters
This revelation from one of AI’s most influential pioneers signals a profound transformation in software engineering that extends far beyond technical implementation. When someone of Karpathy’s caliber—who helped build OpenAI and Tesla’s autopilot—admits to feeling “behind,” it underscores how rapidly AI is reshaping professional work, even for those who created the technology.
The emergence of vibe coding as a mainstream practice represents a fundamental shift from traditional programming to AI-orchestration, where developers increasingly act as prompt engineers rather than code writers. This has significant implications for software education, hiring practices, and career development. Cherny’s observation that new graduates may adapt better suggests potential disruption to traditional experience hierarchies in tech.
For businesses, this signals both opportunity and urgency: companies that successfully integrate AI coding tools could achieve dramatic productivity gains, while those clinging to traditional methods risk falling behind. The “10X productivity” potential Karpathy mentions could reshape competitive dynamics across the entire tech industry, affecting everything from startup velocity to enterprise software development timelines.
Related Stories
- Andrej Karpathy Reflects on ‘Vibe Coding’ Revolution in AI
- The Future of Work in an AI World
- Microsoft AI CEO’s Career Advice for Young People in the AI Era
- CEOs Express Insecurity About AI Strategy and Implementation
- Business Leaders Share Top 3 AI Workforce Predictions for 2025
Source: https://www.businessinsider.com/openai-founding-member-never-felt-so-behind-programmer-2025-12