No Priors Ep. 128 | With DeepLearning.AI Founder Andrew Ng
Andrew Ng discusses the rise of agentic AI, moving beyond scale as the sole driver of progress. He explores how AI-assisted coding is creating a new startup paradigm, shifting the bottleneck from engineering to product management and favoring technical founders. Ng argues for smaller, highly-skilled teams and predicts AI will profoundly empower individuals across all job functions.
Andrew Ng believes the future of AI progress lies not just in scaling models, but in multiple vectors, including “agentic workflows” and novel technologies. He cautions that the public perception of AI has been skewed by the “PR machinery of a handful of companies” focusing solely on scale, while significant advancements are happening elsewhere.
The Rise of Agentic AI
When Ng first introduced the term “agentic AI,” his goal was to define a spectrum of agency rather than a binary classification. This ranges from simple, single-prompt interactions to highly autonomous agents capable of complex planning and reasoning. He wanted to shift the focus from debating definitions to building more capable systems.
The primary obstacle to implementing more agentic workflows today is not a technological deficit but a talent shortage. Ng identifies the key differentiator as a team’s ability to “drive a systematic error analysis process with eval.”
“The single biggest barrier to getting more agentic AI workflows implemented is actually talent.”
Building effective agents requires ingesting proprietary, contextual knowledge that often resides only in the minds of employees, a task that is difficult to automate.
The most impressive and economically valuable examples of agentic AI are currently in software development.
“I feel like bleeding edge of agentic AI I’ve been really impressed by some of the AI coding agents… The second massive bucket of economic value is coding agents.”
He notes that these tools excel at planning and executing multi-step plans, a hallmark of high agency. Other applications, like AI-powered shopping assistants, remain more like demos than production-ready tools.
AI’s Impact on Development and Startups
Ng rejects the term “vibe coding,” arguing it trivializes a “deeply intellectual exercise.” He prefers “AI-assisted coding” or “rapid engineering,” which he describes as mentally exhausting but incredibly powerful for building serious systems faster than ever before.
This acceleration in engineering is fundamentally changing the startup playbook.
- The Product Management Bottleneck: With development cycles shrinking from weeks to days, the new bottleneck is product management—deciding what to build. This makes rapid, gut-driven decision-making, informed by deep customer empathy, more critical than ever.
- The New Founder Profile: In this fast-moving environment, “tech-oriented product leaders” have a significant advantage over those who are less technical. A deep, intuitive feel for AI’s capabilities and limitations is now essential for making strategic decisions. Ng believes this marks a return to the era of highly technical founders like Bill Gates and Steve Jobs.
- The Power of Small Teams: AI tools enable smaller, highly-skilled teams to outperform larger ones by minimizing coordination overhead. Ng is experimenting with this model, observing that “some of the most productive teams I’m a part of now is some of the smallest teams.”
Leadership and Skills for the AI Era
The rapid pace of technological change demands a new mindset. Ng notes that many processes from 2022 are already obsolete. He advocates for leaders and teams to adapt continuously.
One of his core beliefs is that everyone in an organization should learn to code, not to become software engineers, but to communicate more effectively with computers and automate parts of their own jobs.
“When my… assistant general counsel or my CFO or my front desk operator when they learn how to code… they do their job function better because by learning the language of computers, they can now tell a computer more precisely what they want to do for them.”
This shift also changes hiring. An experienced engineer unfamiliar with AI tools may be less productive than a recent graduate who has mastered them. The most effective engineers are those with deep experience who have also embraced AI, putting them in a class of their own.
The Broader Impact of AI
Looking ahead, Ng believes the most significant impact of AI over the next five years will be the empowerment of individuals.
“I think many people will be much more empowered and much more capable in a few years than they are today… People [who embrace AI] will just be so much more powerful and so much more capable than they probably even imagined.”
Just as AI has transformed software engineering productivity, it will unlock similar gains across a vast array of job functions and personal tasks.