5 Questions with Ken Yagen: Turning AI Ambition Into Value with Norwest Kinetic

The constraint on AI-driven growth is surprisingly not tokens or technology. It’s leadership.

That’s the pattern Ken Yagen keeps running into across the Norwest portfolio. At a recent dinner with security and tech leaders, a conversation about AI adoption reinforced an idea he has long held: the best leaders teach a mindset, not a tool.

It’s a principle that has become even more critical as AI changes how companies build and operate. Teams can learn the mechanics of prompting or master the trending AI model or framework, but leaders must rethink how work gets done, where constraints emerge, and how they measure ROI in order to compound the impact of AI.

Ken, Norwest Principal and Operating Executive for Product, AI, and Technology, is hearing versions of these questions across the portfolio. Those conversations shaped Norwest Kinetic, an invitation-only community for Product, AI, and Technology leaders at Norwest portfolio companies. Kinetic engages leadership in four critical ways: portfolio-wide collaboration on benchmarks, half-day Lunch & Learns, immersive bootcamps, and an ongoing peer community for discourse and exchange of best practices.

We sat down with Ken to discuss the challenges leaders are navigating, the mindset required to keep adapting as AI evolves, and how Norwest Kinetic is designed to help them turn ambition into scaled execution and outcomes.

Q.

You work closely with Product, AI, and Technology leaders across the Norwest portfolio. What challenges are you seeing come up again and again?

A.

In our Growth Equity portfolio, the leaders I’m talking to desire benchmark data and want to know:

• Where should we be?
• What are other companies doing that we are not?
• Are we moving quickly enough to stay competitive?
• How do we get from experimentation to something that actually changes how the business operates?

Early-stage AI-native founders are asking about durability of their offerings:

• Are we actually ahead, or does it just feel that way because the space is new?
• What do we own that a model release won’t absorb in twelve months?
• What does it take to make our products reliable enough to sell into an enterprise?
• Do the unit economics hold as we scale, or does inference cost eat the margin?

These look like different questions, but both groups are asking whether their efforts are landing where the constraint actually is. For growth-stage companies the constraint sits in the operating model; for AI-native companies it sits in the product and the moat. The failure mode is the same either way. You can execute perfectly against the wrong constraint and have nothing to show for it two quarters later.

These leaders sometimes just need a thought partner to brainstorm and strategize with. Other times, they’re in the weeds and need an outside perspective to help connect the dots between what’s happening in product, engineering, customer success, and the rest of the organization.

That’s a big part of what led to Kinetic. We have CPOs, CTOs, and AI leaders across a variety of industries and business stages in the Norwest portfolio working through many shared challenges. Kinetic closes the loop. There’s a huge value in bringing those leaders together to learn from practitioners and from each other, and to turn those conversations into ideas and frameworks they can actually put to work.

Q.

What does it actually mean to have an AI mindset, and why does it matter for leaders right now?

A.

I was listening to a recent podcast with Sam Altman, where he shared that he still manually scrolls through his email inbox and does routine tasks the old way, despite having AI tools that could do the work for him. It’s a very relatable anecdote. Habits built over a decade don’t break just because the tool got better and it is genuinely hard it is to change habits that have been ingrained over years. And, organizational habits can be even harder to shift.

But it also underscores why being adaptable is one of the most important qualities a leader can have. You can teach someone how to write the perfect prompt or use a specific coding agent, but those tools are going to keep changing. In fact, prompt-only technical training can sometimes make adoption less effective because people learn a narrow technique without changing how they approach their work.

A recent field study by Gap and Microsoft found that employees who received AI mindset training were twice as likely to produce top-quality work. That research mirrors my own training and experience in Systems Engineering and as a product, AI, and technology leader. This is why mindset development is such an important part of the Kinetic curriculum.

Our goal is to help leaders build the habits and judgment to recognize where AI can create value, experiment within set parameters, learn from the results, and adapt as the technology evolves. If you teach people to think differently about how their work gets done, they become much more effective at adopting whatever comes next.

Q.

Where do you see companies getting stuck as they move from AI experimentation to business impact?

A.

We used to call them vanity metrics back in the day, but I kind of think of it as “AI theater” now. What really matters is outcomes and value delivered.

I encourage our portfolio companies to think about the unit of value for the workstream they’re trying to improve. In customer support, that could be cases resolved, time to resolution, and ultimately customer satisfaction. In product, it’s the value of the feature shipped, how quickly you can get it into customers’ hands and the downstream impact of that feature.

You also have to look at the organization as a system. You can consume a lot of tokens and code really fast (often irrespective of outcome), but you’ve got other constraints in the system. Once you accelerate engineering, maybe product or design becomes the bottleneck or even customer adoption. Ultimately, you want to understand where the constraints are and measure whether the changes you’re making are actually creating a net increase in value.

That’s another reason my systems engineering background has been so helpful. Systems thinking teaches you to step back and look at the relationships between technology, people, processes, and outcomes. That mindset is increasingly important in the work I’m doing with Norwest portfolio companies and in the Kinetic program’s work with leaders.

Q.

How is AI changing what companies need from their product, AI and technology leaders?

A.

Early-stage founders tend to understand that a lot more than established company leaders where these job definitions and responsibilities have solidified and corporate culture has formed.

For product managers, for example, AI creates an opportunity to get back to the core of what a product manager should be doing: understanding the market, understanding the customer, and validating what solutions will deliver value. They can build working prototypes, put them in front of customers, iterate almost in real time, and give engineering something much more focused to build.

That shift is real, and it’s uncomfortable. If a PM can prototype, that changes what a designer does and what a junior engineer does, and people feel that before anyone announces it. Leaders have to hold both things at once. Be assertive about the direction, because the direction isn’t up for debate. Be empathetic about the pace, because you’re asking people to give up skills they spent years building. The displacement risk here isn’t AI taking jobs. It’s the people who don’t get the support to make the transition.

This is again why I think systems thinkers are going to do very well in this agentic software development world. The most effective product, AI, and technology leaders are thinking about architecture, security, data governance, model selection, cost, talent, and how all of those pieces interact.

What AI doesn’t change is accountability. Product still owns whether you’re solving the right customer problem. Design owns the experience. Engineering owns quality, security, and scalability. AI gives each of those functions more leverage, but leaders still need to be accountable for what ultimately gets shipped.

Q.

AI is moving incredibly quickly. What do leaders gain from working through these challenges with their peers?

A.

It can be difficult to know how what you’re experiencing compares with everybody else. I have been in that seat and remember that feeling quite vividly.

Sometimes I can say, “Everyone’s dealing with this. This isn’t just you.” Other times it’s, “You’re way ahead of what I’m seeing elsewhere. Keep it up.” Or I can point to how three other portfolio companies have approached the same problem. That benchmarking and exchange of practical experience is incredibly valuable, and that’s what I want Kinetic to create at a larger scale.

We’re bringing together CPOs, CTOs, and AI leaders who are facing similar decisions, along with practitioners who are actually building solutions today. Every session is designed around something you can act on in the next 30 days. We’re giving them a trusted community and place to work through those challenges together, share what’s working, and walk away with something they can apply. With how quickly this space changes, having that ongoing braintrust can help leaders see around corners they might otherwise miss.