Christoph Nützel

First Principles is a series of unscripted twenty-minute conversations with technology leaders about what leading technical teams actually looks like when the answers aren’t obvious.

Christoph Nützel is CTO at Futurice, a professional services firm working with clients on complex technical transformations across real-time data, agentic AI, and industrial IoT. What follows isn’t a summary of the conversation — the video is the conversation. This is what stayed with me after.

The first question was about the person on his team whose biggest impact this year went unseen by the rest of the organization. Christoph paused. Then he described a principal developer whose work is felt everywhere but recognized nowhere in particular — someone, in his words, “notching at the right right place, the right time, the right person.”

There’s a specific pattern in how senior technical leaders describe their most valuable people. Rarely as headlines. Almost always as verbs and prepositions — where they show up, when they intervene, how they shape the thinking around them. What the organization’s measurement systems can see is the ship date. What Christoph can see is who made the ship date possible. Those two things are not the same, and the gap between them is where a great deal of real work quietly lives.

The second question was about the earliest signal that a team is starting to struggle — before delivery slips. His answer was specific in a way that most answers to this question are not.

“Suddenly an increase in meetings out of nowhere.” Not velocity. Not tickets. Not tone. Meetings piling up, as a leading indicator of things not being solved at the level they should be solved. He also named a second signal, which was harder to say out loud: when he himself starts getting pulled into problems he shouldn’t need to be in. His own calendar, as diagnostic instrument.

Both signals are behavioral. Both precede delivery data by weeks. And both require the leader to be watching the human layer of the organization rather than the delivery layer — which most leaders aren’t, until the delivery layer is already breaking.

The fourth question asked him to be honest about whether AI is actually taking work off his team or adding a new kind of load. He said he couldn’t answer it black and white, and then told two stories.

One was a win. A team using AI to fill in missing data points across a complicated multi-system observability challenge at a client’s sports arena — inference doing detective work that was previously frustrating and slow. “The client loves it, the team loves it.” A real productivity gain in a real high-stakes context.

The other story was harder. His senior technologists reading AI-generated sales proposals sentence by sentence, marking every grammatical error, correcting contradictory statements. What used to be professional expertise going into the proposal directly is now senior expertise correcting AI drafts — which means the same senior people are doing more work, not less, and the register of that work has shifted from creation to correction. He described it as “teaching a fourth grader how a professional company works.”

The pattern is worth sitting with. When AI adoption is measured by tooling metrics — adoption rates, coverage, throughput — the sales-proposal story looks like a success. Adoption is high. The tool is being used. What that measure misses is the specific human cost: a small number of senior people quietly absorbing a new kind of load, one they didn’t have before, in a mode they resent. That load is real. It’s not on any dashboard. And it’s exactly the shape of dysfunction that becomes visible only when someone leaves.

The last question asked what one thing he wishes he could see about his team that he can’t see today. He rejected productivity metrics outright — “you can’t put a number into deep thinking.” Lines of code, pull requests, all of it. He’d seen too many teams gamed by their own measurement systems.

What he wants to see, instead, is something he doesn’t have a clean name for. He called it a “work-life balance score” and then made clear he didn’t mean it in the corporate wellness sense. He meant something closer to a state of the team — whether they are rooted, whether they feel they’re in the right place, whether they can do deep work without the constant background noise of hustle. He said, and this is the sentence I keep returning to: “I don’t believe that constant hustling and doing over hours and running behind that is healthy for a team, nor healthy for the outcomes on time, nor can you really maintain a high quality and throughput all the time.”

He was describing something biological, using the vocabulary he had available. What he wants to measure is the regulated state of the humans doing the work — the difference between a team that can sustain deep work and a team that is compensating with hours. He does it now, he said, by talking to people, by watching how they show up, by trying to read the room. That works at his current scale. It’s a skill that doesn’t scale with the organization.

What this conversation surfaces — invisible impact, meetings as leading indicator, AI redistributing rather than reducing load, a wish for a signal that current instruments don’t measure — is not a summary of what Christoph believes. It’s a snapshot of how one working CTO thinks when the questions don’t have clean answers.

There’s more in the video than there is on this page. If any of what he said resonates with something you’re seeing in your own team, that’s the point of the series.