The New Engineering Stack
What used to take five roles, now fits inside one engineer’s head.
We like to joke about the 10X engineer—but the best ones absolutely exist. And they’re evolving.
This isn’t just about shipping faster. It’s about changing who owns the product vision day to day. In the past, PMs and execs held the context. Engineers executed. But now, the best teams blur those lines. Engineers are shaping UX flows, making architectural tradeoffs based on user feedback, and even adjusting GTM timing based on real-world system performance. They’re compounding product judgment across every layer of the stack—from model tuning to onboarding UX.
In the old model, ownership was clearly divided:
PMs wrote the specs
Designers owned the experience
Engineers executed the tickets
Business execs made the calls
Data scientists ran the analysis
Ops managed reliability
But in today’s AI-accelerated world, those boundaries are blurring fast.
The best engineers I know are learning to move differently—not just faster or louder, but with sharper judgment and a much broader sense of ownership. Because of what AI tools now enable, the pieces of the stack that were once siloed across roles are being rebundled inside a single person. One engineer can now prompt, scaffold, ship, and refine an entire product experience. That shift changes everything.
That doesn’t mean those roles go away. Designers, PMs, pricing specialists—they’re all becoming more powerful, too. In fact, many can now prototype and ship full applications themselves. But what’s shifting is that the bar for contribution is rising across the board. Before, siloed skillsets were prized. Now, the ability to move fluidly across domains—whether you’re an engineer who can shape product, or a PM who can ship code—is becoming the true differentiator. Great teams won’t be one-person bands. But the people who thrive on those teams will be deeply cross-functional.
There used to be successful founders who only spoke to computers well, or those who only spoke to people well—coders vs. salespeople. Today, being a well-rounded athlete can serve you better. Even on small, sharp teams, the push is toward being more cross-functional, not just pointy in one area.
They’re not just shaping features. They’re deciding what not to build. They’re pruning scope, modeling UX friction, stress-testing architecture for scale, and asking: will this hold up over time, or should we rethink the frame?
And yes—they’re weighing in on hiring, GTM timing, and even pricing models, because their visibility into how the product performs in the real world gives them a privileged vantage point. They’re not just taking tickets. They’re rebuilding the factory.
From spec takers to synthesisers
They don’t just take a ticket and ship it. They ask: Why now? Who does this help? What’s the downstream effect? They reframe the work based on systems thinking and product impact.
From builders to judgment amplifiers
They choose which problems are even worth solving. They say no—often. They’re the ones who delete features, merge workflows, and optimize for clarity.
From ship velocity to consequence velocity
The real metric isn’t “how fast can we launch.” It’s: how long until this thing makes us smarter, faster, more focused? These engineers optimize for compounding advantage, not feature count.
From rigid tools to fluid tools
They move across layers of abstraction—prompting, scripting, editing AI agents, shaping interfaces—without getting stuck in job titles or tech silos. The tools don’t define their role; the outcome does.
From generalists to signal-maximizers
They don’t build to impress. They build to learn. Every commit, prompt, or interaction is a chance to generate sharper signals—for users, for the model, for the team.
So what should teams do differently?
For engineers:
Cultivate a product POV—ask more “why” questions.
Learn to use AI tools without being defined by them.
Design with iteration loops in mind—what makes the system smarter over time?
Don’t just code—synthesize.
For product leaders:
Hire for judgment, not just velocity.
Reward deletion, simplification, and insight—just as much as shipping.
Invite engineers into product decisions early.
Shift from roadmap gatekeeping to roadmap curation.
This isn’t the end of software engineering. It’s the beginning of a deeper, sharper, more judgment-driven era.
In an era where tools are cheap and output is infinite, it’s judgment that compounds. And the best engineers I know? They’re quietly mastering it.