Thanks so much for joining us today. We have over 20,000 participants registered for this event—both longtime learners and newcomers. I believe you’re here because, like me, you sense we’re not witnessing the end of programming, but its remarkable transformation. In over 40 years in the software industry, I’ve never seen a moment as exciting as this one.
I organized this event out of growing frustration with a familiar narrative: that AI will replace programmers. Variations of this prediction resurface with every major technological advancement—and each time, it misses the mark. Not just a little, but fundamentally. It misunderstands how technology actually evolves.
Programming is, at its core, a conversation with computers. It’s how we turn human intention into machine action. History shows us a continuous trend of building better interfaces for this conversation—from physical wiring to assembly code, from high-level languages to the web, which integrated backend logic into readable documents. Today’s large language models (LLMs) are simply the next evolution, making computational power more accessible and human-friendly than ever.
And every time the barrier to communicating with machines gets lowered, we don’t end up with fewer programmers—we unlock entirely new areas where software can have an impact.
There’s a pattern: a breakthrough resets expectations, sparking intense innovation, which is eventually followed by consolidation and a period of relative calm—until the next disruption redefines the landscape once again.
The Historical Pattern of Expansion
Think about how far programming has come. Early developers had to physically wire circuits. Then came the stored-program model, followed by binary inputs through front panel switches, then assembly language, then high-level programming languages. The World Wide Web transformed computing once again, making it feel conversational and accessible through HTML and links that triggered backend logic.
With each leap, skeptics worried “real programming” was dying. But the opposite happened. Programming expanded into new domains, attracting more people and creating new specialties.
Take the digital spreadsheet—VisiCalc, created by Dan Bricklin and Bob Frankston. Initially prototyped in BASIC, it had to be rewritten in assembly for performance reasons. This example shows the value of rapid prototyping tools paired with deep technical expertise for production-level systems.
Decades later, Tim Berners-Lee’s creation of the web on a NeXT machine opened up programming to millions. Many learned by simply viewing and editing HTML in their browsers. Countless billion-dollar companies were born from this grassroots tinkering.
AI-Assisted Programming: Democratization on Steroids
That same pattern is unfolding again—but on a vastly larger scale.
A tech executive recently shared how his high-school-aged daughter interned with a Stanford biomedical professor. With no prior coding experience—her interests were in biology—she was asked to explore a problem: traditional pulse oximeters are unreliable, and the professor wondered whether retinal capillaries could be used instead. She used ChatGPT to process retinal images, isolate capillaries, and write code for analyzing oxygen saturation.
This project, which once would’ve required a grant, a hired team, and months of research, was now initiated by a curious intern. What this really shows is that the cost of experimentation has collapsed—while the scope of what we can explore has exploded.
But that prototype is only the beginning. Creating a reliable, medically approved product still demands professional engineers who understand robust system design, regulation, testing, and scalability.
Right now, many are focused on improving existing solutions with AI. The real breakthroughs will come from solving problems we once thought were impossible or impractical.
The New Spectrum: From Vibe Coding to AI Engineering
We’re seeing the emergence of a new continuum of programming. On one end is “vibe coding”—fast, intuition-led development aided by AI. On the other end is structured AI engineering—rigorous system design using AI as a component.
It’s similar to how the early web evolved from basic HTML to complex ecosystems with frameworks, APIs, and cloud infrastructure. That shift didn’t reduce the need for developers—it multiplied it. We saw the rise of frontend and backend engineering, DevOps, security, and more.
Now, we’re seeing the same with LLMs and AI agents. The base model is just the start. The real work lies in integrating, tuning, and embedding these models within larger software systems.
The New Hybrid Computing Paradigm
Applications powered by AI—whether it’s a chatbot or a full product—are hybrids, blending traditional software with generative intelligence. Tools like ChatGPT or Perplexity are just the engine. Around them are workflows, interfaces, and logic shaped by designers, PMs, and engineers.
As one developer noted, the model is like a car engine. You still need everything else—chassis, tires, transmission—to make a sports car. And today, building that car requires orchestrating two very different types of computers: one creative but fallible, the other precise but rigid.
Another technologist put it this way: coding is reliable but inflexible, while AI is flexible but inconsistent. The new developer’s job is to write “metacognitive recipes”—code that manages and shapes AI inference. Doing this well can turn vague outputs into reliable tools.
This is the new terrain. We’re not at the end of programming—we’re at the beginning of its most profound reinvention.
A Renaissance of Innovation
It’s a thrilling time for software. After years of incremental progress, we’re experiencing a creative renaissance. We’re not just writing code faster—we’re redefining what software is, who can create it, and what kinds of problems we can now solve.
Throughout this event, we’ll explore three key dimensions of this transformation:
- How to partner effectively with AI to supercharge your workflow
- Best practices and pitfalls in building production-grade AI systems
- New opportunities unlocked by AI-driven exploration and experimentation
Software development was getting predictable. Now it’s anything but. The excitement is back, and so is the possibility.
As we dive into these sessions, I encourage you to ask yourself: What challenge that seemed impossible yesterday might now be within reach?
Let’s embrace this moment as pioneers—not with fear, but with curiosity, boldness, and creativity.
