Nvidia CEO Jensen Huang said AI is changing software engineering by moving developers away from repetitive coding. He said engineers are now building agents, systems, and tools that could automate complex work across the workplace.
In an interview published by Nvidia on Wednesday, Huang said “agentic systems” are creating a new category of technical skills. He said many software engineers are now working on agents instead of routine coding tasks.
AI Agents Redefine Software Engineering Roles
Huang said agentic systems are creating new technical skills, with many software engineers now focused on building agents. He added that Nvidia’s engineers prefer working on agents instead of writing Python code.
Huang said AI is reducing the amount of manual coding engineers need to perform. That gives them more time to design AI systems, build benchmarks, and create safety guardrails.
The Nvidia CEO said developers are now trying to move routine work into AI agents. He described the process as technical and creative. “You’re taking all the mundane work, and you’re trying to get this agent to do it.”
He added that the shift requires imagination, creativity, and technology. His comments point to a broader change in software development, where engineers guide automated systems rather than only write code line by line.
Huang has often promoted AI assistants in the workplace. Nvidia has also said it plans to expand the use of AI agents across its divisions to improve productivity.
Huang Rejects Extreme Claims About AI
The Nvidia CEO also addressed wider fears about AI during an earlier appearance on The All-In Podcast. He said AI should not be described as a biological being, an alien force, or a conscious entity.
Huang described AI as computer software. He also pushed back against claims that people do not understand the technology at all. “We understand a lot of things about this technology,” he said.
He also discussed Anthropic, saying the company’s technology is “incredible.” Huang added that Nvidia is a large consumer of Anthropic technology.
Huang said warnings about AI capabilities could be useful. However, he said scaring people is less useful because technology is too important. Huang said technology leaders should be more careful when speaking about AI risks.
He highlighted that leaders need more humility when discussing the future of AI. Huang added that no one could fully predict how the technology would develop.
However, ARK Invest CEO Cathie Wood also reacted to Huang’s comments. She said she agreed with him that government regulation is the risk.
I agree with Jensen that government regulation is the risk, and wonder why developers say they don’t understand the results of the software they are building. They should explain what they mean. https://t.co/IukPc15Rdj
— Cathie Wood (@CathieDWood) July 9, 2026
Wood also pushed back on claims from developers who do not understand the results of the software they are building. In her view, those developers should explain what they mean.
Together, Huang’s remarks present AI as a major shift in software work, not a replacement for engineering judgment. His view places developers closer to system design, automation strategy, and responsible deployment.
Related: Swift Launches Blockchain Ledger Pilot With 17 Global Banks
