Nvidia CEO Jensen Huang wants engineers to stop coding and start thinking; here’s why

Jensen Huang, Nvidia CEO

• Nvidia CEO wants engineers to spend zero percent of time writing code
• Huang says AI should handle coding so engineers focus on problem-solving
• Developers warn AI-generated code may cause reliability and technical debt issues.

Did our AI summary help?

Jensen Huang, CEO of Nvidia, has sparked debate in the software industry after saying he wants engineers to spend “zero percent” of their time writing code. Instead, Huang wants them to focus on identifying and solving problems that have not yet been addressed, while artificial intelligence handles most of the coding work.

The remarks were made during recent interviews and public discussions where Huang outlined how AI tools are changing the nature of engineering work at Nvidia.

Why Nvidia wants less coding

According to Huang, coding itself is a task, not the end goal. The real purpose of engineering, he argues, is problem discovery and problem-solving. Nvidia engineers now use AI coding assistants extensively, allowing them to offload syntax-heavy work to machines. This, Huang believes, frees human engineers to think more deeply about system design, algorithms, and unsolved challenges.

He has described this approach as “purpose versus task,” where AI handles repetitive execution while humans focus on intent and direction. In Huang’s view, this shift improves productivity and leads to more meaningful innovation rather than simply faster code output.

The radiology comparison

To explain his thinking, Huang often points to the field of radiology. Years ago, experts predicted that AI would replace radiologists by reading scans faster and more accurately. Instead, the number of radiologists has increased. The reason, Huang says, is that image reading was only one task within a broader purpose: diagnosing disease and improving patient outcomes.

By removing routine work, AI expanded what professionals could do. Huang believes software engineering will follow a similar path, with AI-written code enabling engineers to tackle more complex and higher-level problems.

Concerns from the developer community

Not everyone agrees with Huang’s outlook. Some AI tool builders and developers warn that relying too heavily on AI-generated code without proper review can introduce technical debt and reliability issues. Studies have also suggested that AI coding assistants do not always improve productivity for experienced developers.

These concerns highlight a gap between AI’s promise and its current limitations, especially for large, complex systems.

Huang’s vision does not eliminate engineers, but it does redefine their role. If AI continues to improve, engineers may spend less time typing code and more time setting goals, validating results, and making critical decisions. Whether this shift benefits the wider developer workforce remains an open question, but it signals a clear direction for how Nvidia sees the future of engineering work.