There’s something about the smell of fresh croissants, strong coffee, and the buzz of twenty-four technical leaders squeezed into one room that sets the stage for a certain kind of magic. Not the scripted, predictable kind you get at industry conferences with slide decks and polite applause, but the real thing: laughter when someone cracks a pointed joke, raised eyebrows when a bold claim lands, and a back-and-forth that can only happen when everyone knows they’re talking to peers who actually get it. That was the scene this morning at the AWS offices in Amsterdam, where the CTO Network gathered once again.

From the moment people walked in, the energy was clear. This wasn’t going to be a passive morning of keynotes and note-taking. The CTO Network is built differently: no sales pitches, no safe consensus. Just raw conversations with people who wrestle with the same challenges every day — building, scaling, securing, and sometimes fighting with technology that moves faster than anyone’s governance frameworks can keep up with.

The keynote: A wave as big as the cloud

Our opening session was led by Ahmed Bashir, CTO of DevRev and former Apple engineering leader who helped shape products like iMessage and Siri. Ahmed doesn’t do small talk, and his keynote was a reminder of why the current AI wave feels like the most significant shift since the cloud and mobile revolutions of the early 2010s. He laid out how nearly every developer today — 92% of them — already leans on AI coding tools like Copilot or ChatGPT. But instead of celebrating that fact, he posed a harder question: what does it mean when enterprises are freezing headcount growth, keeping it flat at 3–5%, and diverting those budgets into AI subscriptions costing $50 to $200 per employee?

Ahmed described DevRev’s approach as not simply building AI tools, but creating an enterprise experience that feels as seamless and transformative as the first iPhone did for workplace hardware. He spoke about the power of context over sheer volume — how only a sliver of data, maybe 5%, truly matters when you’re trying to connect the dots between tickets, customers, and teams. He explained how they’ve architected privacy from the start, storing data in European data centers and distinguishing between transactional data accessed via APIs and contextual data brought into memory. His point was simple, but powerful: the way you design data, privacy, and governance into your systems today determines whether your AI transformation will be sustainable tomorrow.

The highlight came when he ran a live demo. On one screen, DevRev’s Cloud Code engine answered a complex question in six seconds flat, pulling in 367 sources and summarizing themes across customer complaints with clear takeaways for leadership. On the other, a standard AI system was still grinding away at 45 seconds, with little to show. It was a visceral illustration of what context-aware AI can achieve, and you could feel the shift in the room as CTOs leaned in.

The roundtables: Where sparks fly

If the keynote set the stage, the roundtables lit the fire. We split into three groups, rotating every twenty minutes, to get into the gritty details: AI security, product management, and implementation best practices.

At the AI & Security table, moderated (or rather refereed) by Builders’ own Michael van Lier, the discussion took off immediately. On one side sat the pragmatists: “ship fast, security is baked in.” On the other side, the absolutists: “zero-trust forever, review every line, even the AI’s.” It wasn’t long before the conversation cut to the heart of the tension every CTO feels today: balancing the speed of innovation with the weight of responsibility.

Stories poured out. One participant revealed how their security team had successfully used AI to reverse engineer an obfuscated SDK in just two weeks — a task that normally takes months. Another described how their company requires 100% test coverage on any AI-generated code before it’s accepted into production. A third shared their “extreme” approach of giving AI agents their own identities with limited permissions, running development inside isolated virtual desktops to prevent any accidental leaks. Some in the room considered it overkill, others saw it as the only sane way forward.

The conversation circled back, again and again, to one looming problem: memory. AI systems don’t forget. What happens when an AI model “remembers” sensitive data that governance rules no longer allow? What does it mean to programmatically erase knowledge at scale? No one had a good answer, and that silence spoke louder than anything else.

Why it matters

By the time the sessions wrapped, the coffee was cold and the croissants long gone, but the debates were still alive in everyone’s minds. That’s the point of the CTO Network. It’s not about agreeing, it’s about sharpening each other’s thinking. And in today’s discussions, a few truths surfaced clearly: AI adoption is already near-universal, but governance is miles behind. Enterprises are spending freely on tools without knowing how to measure their return. And the companies that will win are the ones that stop treating AI as a black box and instead design architectures that embed context, privacy, and auditability from day one.

There’s also something reassuring about the friction itself. When opinions clash this starkly, it’s a signal that the field is still wide open — which means opportunity. Somewhere between “ship fast” and “zero-trust forever” lies the next great security playbook, and someone in this network may be the one to write it.

Closing thoughts

Walking out of the AWS offices, the conversations carried on down the street. Some CTOs debated compliance frameworks, others swapped stories about AI in their own organizations. That’s the mark of a successful morning: nobody wanted to stop talking.

The CTO Network isn’t about consensus. It’s about leaving with new questions, sharper instincts, and maybe the next big idea. See you at the next one.