The Rise of AI-Powered Coding: Are Developers Becoming Managers?

The software development landscape is undergoing a radical transformation, driven by advancements in artificial intelligence. While the image of developers hunched over lines of code remains a common one, a growing number of tech companies are leveraging AI to automate significant portions of the coding process. From Spotify to Google and Anthropic, engineers are shifting their focus from writing code to design, management, and critical judgment, raising questions about the future role of the software developer.

AI Takes the Reins in Code Generation

The shift towards AI-assisted coding isn’t a distant prospect—it’s happening now. Gustav Söderström, co-CEO of Spotify, revealed that the company’s top engineers haven’t written a single line of code since December, relying instead on an internal AI system called “Honk.” TechCrunch reports that Spotify shipped over 50 new features and changes to its streaming app throughout 2025, many of which were developed with the help of AI. Similarly, Anthropic reportedly utilizes AI to write 70-90% of its code, and Google leadership stated in October that AI agents are responsible for half of all code written at the company. The Economic Times

How AI Coding Systems Work

Systems like Spotify’s “Honk” leverage generative AI, specifically Anthropic’s Claude Code, to accelerate development and deployment. Engineers can instruct the AI via platforms like Slack—even while commuting—to fix bugs or add new features. The updated code is then delivered back to the engineer for review and integration into production. Storyboard18 This process dramatically speeds up coding and deployment timelines, allowing companies to release new features more rapidly.

The Evolving Role of the Software Developer

As AI handles more of the coding tasks, the role of the software developer is evolving. Instead of focusing on programming languages like Python or JavaScript, developers are increasingly valued for their ability to design software at an architectural level, anticipate potential problems, and exercise critical judgment. Julian Togelius, a professor of computer science and engineering at New York University, notes that developers are moving away from syntax and towards design and management. This transition requires a different skillset, one that emphasizes people management and the ability to oversee multiple AI coding agents.

The Challenges of AI-Assisted Development

While AI-assisted coding offers significant benefits, it also presents new challenges. Managing multiple AI agents can be demanding, potentially leading to burnout as developers constantly monitor and adjust the AI’s output. Togelius points out that this constant oversight can create a disconnect, where developers may *sense* more productive but aren’t necessarily *being* more productive. The rapid pace of technological change requires ongoing training and adaptation for engineering teams. Google addresses this by embedding employees within engineering teams to stay current on new tools and capabilities and facilitate knowledge sharing.

Beyond Code: AI’s Expanding Role in the Software Lifecycle

AI’s impact extends beyond simply writing new code. Developers are also using AI to create test cases, analyze data, and debug software. TechCrunch reports that this trend is accelerating, with a 2025 Doraa Google Cloud research program survey finding that 90% of software development professionals were using AI at work as of September – a 14% increase from the previous year. Looking ahead, companies like Google are focusing on using AI to maintain and scale applications once they’ve been deployed, marking the “next frontier of AI” in software development.

Key Takeaways

• AI is rapidly automating code writing, with some companies reporting engineers haven’t written code manually for months.
• The role of the software developer is shifting from coding to design, management, and critical judgment.
• Managing AI agents presents new challenges, including potential burnout and the need for continuous learning.
• AI’s impact extends beyond code generation to encompass testing, data analysis, and application maintenance.