In today’s FemalExperts Role Model Interview, we’re pleased to introduce Sarah Gottwald. She is a founder, executive, and expert in strategy, digital transformation, and artificial intelligence. After nine years at BMW and three years as the managing director of a VC accelerator, she launched her own company six months ago. Through deAI Labs and Mira OS, she is now dedicated to helping companies successfully implement artificial intelligence in collaboration with their employees.
In an interview with FemalExperts, she talks about her personal and professional journey, her transition from a large corporation to starting her own business, leadership and change, and female empowerment in the tech and AI worlds.
1. Sarah, when you look back on your career so far, what experiences or decisions have had a particularly profound impact on you and led you to continually explore new career paths?
Looking back, it wasn’t so much about individual decisions as it was about three values. The first is growth: When I realize I’m no longer learning anything new in a role, I get restless. The second is making a difference myself. I don’t just want to analyze and make recommendations; I want to see that my work actually brings about change. And the third is independence: being free in what I do and how I do it.
Every step I’ve taken in my career can be traced back to one of these three points.
Whenever someone felt left out, I made a change. From the outside, it sometimes looks like a break with the past, but to me, it was a logical step.
2. You worked at BMW for nine years, first in finance and later in strategy, partnerships, and M&A. What did you learn during that time about your career, leadership, and perhaps even about yourself?
My most important insight on the topic of careers is an uncomfortable one: Performance is a prerequisite, but it’s not enough. It’s true that without professional expertise and genuine dedication, you won’t get anywhere in a corporation. But the system doesn’t automatically reward that performance. There’s no mechanism that reliably identifies who’s doing a good job and promotes that person to higher positions.
It’s the people around you who make the difference. Someone with a manager who challenges and supports them has different opportunities than someone with the same performance level but without that support. It’s not fair, but it’s the reality. It’s worth realizing this early on.
Through leadership, I’ve learned just how challenging it is within a corporate group. There are always employees who really want to make a difference. And there are those who are constantly dissatisfied, but whom you still have to bring along. Managing both at the same time is harder than it looks from the outside.
One moment in particular has stuck with me: an assessment center. Many of the questions remained abstract because I was asked to give examples that didn’t feel authentic. Only after I became a managing director myself would I have been able to truly answer them. Looking back, I also realize that a corporate group is a very well-protected system. Many people have lost that sense of gratitude.
And I’ve learned about myself that I’m different from many people around me in one respect. I have a strong desire to create: to build something and decide of my own free will which direction to take. That was the sign that I would eventually do something on my own.
3. After nine years at a large corporation, you decided to quit. Do you still remember the moment when it became clear to you that you wanted to do something different? And how hard was it to leave behind the security of an established company?
I remember that very well. A friend said something to me back then that has stuck with me: “Don’t always focus on the downside—everything that could go wrong—but on the upside—everything that could change for the better.” It sounds simple, but it completely changed my perspective. We carefully assess risks, but we rarely examine opportunities with the same level of scrutiny.
There’s another moment that has stayed with me. I had handed in my resignation, and my HR manager—who knew me reasonably well—called me and said she could just throw the resignation letter away. One sentence, one phone call, and everything would have gone back to the way it was before.
That hit me harder than I expected. Nine years, a steady paycheck, an environment where I knew how everything worked, and people I liked. All of that was suddenly up for grabs again, and I had to let it go a second time. Letting go of that security was the real decision.
4. After that, you spent three years as the managing director of a VC accelerator, where you got to know companies and founders from a completely different perspective. How did that experience change your view of entrepreneurship and successful startups?
During that time, I truly understood for the first time just how difficult it is to make money. In a large corporation, revenue is just a number in the financial reports. As a managing director, the question is whether salaries can be paid at the end of the month.
At the same time, I’ve seen how much the broader conditions influence whether someone even decides to start a business. In Germany, the path to employment is better established than the path to self-employment—in terms of taxes, social benefits, and culture. In the 2025 German Startup Monitor, nearly one in three founders says they would start a business abroad today. 88 percent cite less bureaucracy as the reason, while 71 percent cite better access to capital. And according to Bitkom, one in four German tech startups is considering leaving Germany because there is too little capital available here.
I find it interesting that the U.S. is chosen less often as a place to start a business. Many are now looking to other European countries. To me, that makes the situation even more urgent: We’re not losing out to an all-powerful Silicon Valley, but to our immediate neighbors.
Complaining about bureaucracy is, of course, a sweeping generalization—I’m well aware of that. It’s the kind of thing everyone says, and that doesn’t make it any more accurate. Still, I can only agree. It’s not about individual regulations, but about the total amount of time involved. That time is missing where the success of a young company is decided: with the customers.
[Sources: German Startup Monitor 2025 (Startup Association/PwC, September 29, 2025, n = 1,846): 29% would choose to launch abroad if they were to start a company again; 88% cite less bureaucracy, and 71% cite better access to capital. Bitkom Research, “Lack of Capital Drives Startups Abroad” (March 10, 2026, n = 133): 25% are considering moving abroad due to a lack of capital.]
5. Today, you’re a founder yourself. What ultimately made you decide to stop just supporting other companies and innovations and instead build something of your own with deAI Labs and Mira OS?
I spent years advising founders, but at some point I realized that I had never actually experienced many of the things I was talking about. I could explain what mattered without ever having been in a situation where it really became a matter of survival.
But something else was the deciding factor. I’m convinced that building a business as an entrepreneur is one of the most difficult challenges one can take on. And that’s exactly what I was determined to try for myself. Not because I was certain it would succeed, but because I didn’t want to live with the question of whether it could have succeeded.
The fact that it has become an AI topic, of all things, also has to do with an observation I’ve made. Many people are afraid of AI or of the changes it will bring. I take a different approach: I’d rather be at the forefront of this topic than not understand it at all. Those who understand how this technology works and where its limits lie can shape it.
6. With Mira OS, you’re tackling a topic that’s currently on the minds of nearly every company: artificial intelligence. What specific problem do you hope to solve with Mira OS, and how does your approach differ from other AI solutions?
Many AI projects fail not because of the model, but because of the data. Gartner predicts that, by 2026, companies will abandon about 60 percent of AI projects that lack an AI-ready data foundation. This aligns with what we see among our clients. The knowledge is there—in ERP systems, databases, documents, and tickets. It’s just that it has accumulated in a way that neither humans nor AI can make use of it.
A recent Alteryx survey of 1,400 technology leaders worldwide shows the same thing from a different perspective: 77 percent say that business context is crucial for accurate and relevant AI results. At the same time, 53 percent say their organization fails to translate this context into their systems and workflows. Our business model lies between these two figures. The bottleneck isn’t the model—it’s the context.
We extract this data largely automatically and use it to build a controlled corporate context—in other words, a corporate memory that actually works. Every response is cited with its source and takes the respective role into account: Management sees different information than the accounting department. The system is hosted in the EU, is GDPR-compliant, and can be hosted on your own infrastructure upon request.
For me, the difference lies in the depth. Databases are often the real challenge: they are large, opaque, and practically never standardized. Typically, the access logic is written by hand for each system. Our systems automatically map databases. Building on this, evolutionary algorithms learn the structure, create their own query functions, and continuously optimize them. Added to this are aspects that are easily underestimated: a robust role- and permission-based model, complete source citations, and properly implemented data protection.
I often put it this way in conversations:
Many AI systems operate like an army of highly motivated interns. That works for a while. Once they reach a certain size, however, they become unreliable, simply because they lack experience.
[Sources: Gartner, “Lack of AI-Ready Data Puts AI Projects at Risk” (February 26, 2025): 60% of AI projects without an AI-ready data foundation are expected to be abandoned during 2026. Alteryx, “2026 IT Leader Research” (August 13, 2026, n = 1,400): 77% cite business context as critical; 53% have difficulty incorporating it into AI systems.]
7. In small and medium-sized businesses in particular, new technologies often come up against established structures, existing processes, and a wide range of experiences with digitalization. In your view, what are currently the biggest challenges in implementing AI?
The first challenge is the state of the data. In most companies, data has accumulated over decades, spread across different systems and often without documentation. Frequently, the critical knowledge isn’t stored in a system at all, but resides in the minds of experienced colleagues—and is lost bit by bit with every retirement.
The second challenge is negative past experiences. Many small and medium-sized businesses have already tried something: a chatbot that provided inadequate or even incorrect answers, or a digital transformation project that fizzled out. This disappointment runs deep. It is the real obstacle you’re working to overcome—not the technology itself.
The third challenge is the lack of measurability. Time and again, we encounter areas without reliable metrics: no processing times, no first-resolution rates, no performance analyses.
The fourth challenge is control. For a medium-sized company or a public sector partner, the question of where the data is stored and who has access to it is no minor issue. It is a prerequisite for a project to even get off the ground.
8. In particular, you have experience guiding teams through change processes, even when previous digitalization projects have already failed. How do you build new trust and foster a genuine willingness to embrace change in such a situation?
Every project is unique, and trust can only be earned on a personal level. You have to understand people, grasp how the company works, and respond accordingly. That can’t be standardized.
In practical terms, this means, above all, listening carefully—not just to what is said, but also to how it is said. That’s where the real information usually lies: where the frustration stems from, where someone is worried about their role, or where a promise was once made but not kept.
That’s why we never start with technology alone. At one of our clients—a software company for local governments—we first spent a day documenting the day-to-day work in their support department: Where do inquiries come from, how are solutions sought, and where is time wasted? Up to 400 tickets are processed there every day, and the database contains 111,000 historical cases. Nevertheless, the answers were primarily stored in people’s minds.
And you have to make yourself accountable. Before the go-live, we worked together to define a test catalog that will be used to validate the system. In doing so, we contractually guaranteed a certain level of answer quality. Every answer is backed by its source. That’s exactly what builds trust—not the promise that it will work this time.
9. You’ve now experienced a wide variety of environments: a large corporation, venture capital, executive management, and starting your own company. Has your understanding of good leadership changed along the way, and if so, how?
Very much so. Leadership means something completely different depending on the context. In a corporate setting, it’s about getting everyone on board and making the most of existing structures.
As a managing director—and now as a founder—the goal is almost entirely about attracting the best people for the open positions. You can’t bring along anyone who doesn’t want to come along. There’s simply not enough time or money for that.
In a small business, leadership begins with the very first job interview. If I can fill a role with the right person, everything that follows becomes infinitely easier because I can trust that person. The smaller the team, the more crucial each individual is. I can only build something with people who are fully committed.
After that, it’s about understanding people’s interests. Why does someone do something a certain way? What drives them, and what holds them back? For me, learning to understand people is one of the most important leadership skills of all. Added to this is a high degree of personal responsibility, guided by a clear rule: exactly one person is responsible for each issue—not several. As soon as responsibility is spread out, in the end no one feels accountable.
We also work remotely for the most part, and that changes leadership yet again. It no longer works through presence or control. What remains is motivation through the content: through the topics themselves, through a shared belief that we can truly make a difference, and through the desire to succeed. If that’s not there, even an office won’t help.
Resilience is a major issue because setbacks happen time and again. You have to guide your employees through difficult times and motivate one another. As a leader, you experience the same setbacks, yet you still have to be the one who sets the course for the next step. In doing so, you mustn’t lose sight of who you are.
The most important insight, however, is something else entirely. Working within a predetermined structure can be highly complex in terms of content. Nevertheless, it is something entirely different to independently consider what the right next step is to ensure the company’s growth and employees’ salaries. The crucial issue is focus: What is the most important thing I need to do next? For me, this boils down to a question I ask myself time and again in my daily life: How can I create value? Everything else is just a distraction. And then it’s a matter of setting priorities and sticking to them. The second part is often harder than the first. This way of thinking is hard to train yourself to adopt; you have to experience it and learn it.
10. The tech, AI, and investment worlds remain predominantly male-dominated. How do you experience this as a founder and executive? Are there situations in which you particularly notice that women still face different hurdles in these fields?
I don’t usually experience it as outright rejection, but rather in the form of unspoken assumptions. There’s a survey that illustrates this. About 1,000 people evaluated the same AI-generated resume; only the first name was changed. In one case, the person’s name was James; in another, Emily. Everything else was identical. The difference was particularly clear among male respondents from Generation Z: 97 percent rated James’s application as strong, while 76 percent did so for Emily’s. Overall, respondents were 22 percent more likely to doubt Emily’s trustworthiness and twice as likely to doubt her competence.
These results are so revealing to me because, even with identical résumés, it’s clear that performance wasn’t the only deciding factor.
To be honest, though, I feel very comfortable in my social circle and am glad to have strong people around me who value me and don’t fundamentally question my abilities.
[Source: Code For Good Now, “The AI Judgment Penalty” (April 2026, n = 1,000, United Kingdom): Doubts about Emily’s trustworthiness were expressed 22% more frequently, and doubts about her competence were expressed twice as often. The figures 97% and 76% refer to male respondents from Generation Z. Proprietary, non-peer-reviewed survey.]
11. For FemalExperts, female empowerment means not only talking about equality, but also supporting women in making their expertise, their voices, and their influence visible. What does female empowerment mean to you personally, and what concrete steps can women—as well as companies and networks—take to ensure that more women play a leading role in technology, innovation, and entrepreneurship?
For me, female empowerment means, above all, the power to shape things: sitting at the tables where decisions are made and actually having a say there. Not visibility for its own sake, but influence over decisions, products, and capital.
In my experience, for women, this means showing what you’re capable of early on and not waiting until you feel 100 percent qualified. You should seek out people who challenge and support you, and think in economic terms—that is, talk about salary and equity—rather than hoping that your good work will be noticed.
A friend once said to me:
If you don’t ask for what you want, you won’t get what you want.
The same applies to companies and networks, only in a more concrete way: don’t just invite women—bring them onto the stage and into decision-making bodies. When making hiring decisions and investments, consider whether the focus is on risk or potential. Mentoring should not be seen as a friendly gesture, but as targeted support. When it comes to concrete access, visibility, and opportunities, sponsorship is also needed.
12. As we wrap up our role model interviews, we always like to look ahead: What is your vision for your own future and for Mira OS, and what advice would you like to share with women who are currently facing a major career decision or thinking about starting their own business?
For Mira OS, the vision is clear: We want to build the data and context layer on which European companies can build their AI applications—in a sovereign, transparent manner and using their own data. If, in a few years, a small-to-medium-sized business says that its corporate knowledge has been made accessible and that every AI-generated response can be verified against original sources, then it will have been worth it. It’s important to me that this infrastructure is built in Europe.
For me personally: I want to be and remain an entrepreneur. I want to continue building something that wouldn’t exist without me, and never stop learning along the way.
And the message I’d like to share: Free yourself from the opinions of others and believe in yourself. There’s always a solution—sometimes you just can’t see it yet. And on a practical note: You don’t have to know how to do everything beforehand. You grow precisely through the steps you’re most intimidated by.
Dear Sarah, thank you so much for taking the time to share your experiences and perspectives with us and our FemalExperts community. Your journey shows that professional development doesn’t always have to follow a straight path, and that it’s precisely the courage to step outside familiar structures that can open up new opportunities. We wish you and your team every success in the continued development of Mira OS, and we’re excited to see what the next chapters will bring.
About the author
Kinga Bartczak advises, coaches and writes on female empowerment, new work culture, organizational development, systemic coaching and personal branding. She is also the managing director of UnternehmerRebellen GmbH and publisher of the FemalExperts magazine .
