I recently went one-on-one with Sam Dorison, co-founder and CEO ReflexAI.
Adam: Thanks again for taking the time to share your advice. First things first, though, I am sure readers would love to learn more about you. How did you get here? What experiences, failures, setbacks, or challenges have been most instrumental to your growth?
Sam: My family has always had a core value of contributing to communities. After graduate school, one way that I lived this value was volunteering with a crisis line for several years while also working at McKinsey. Regardless of my travel schedule, every Sunday night included a three-hour shift as a crisis counselor. I ultimately left consulting to work full-time at the crisis line. From that vantage point, I could see a major item that was holding us back: new counselors needed extensive practice before they were ready to take real calls, and the bottleneck was always this part of training because it was the most manual. We had already made huge advances for the knowledge components via an LMS, but manual practice was the major, manual issue.
My ReflexAI co-founder John Callery and I led the teams at The Trevor Project, developing a generative AI simulator that gave trainees the practice they needed and deserved. These tools launched in 2020, over two years before the launch of ChatGPT. And in 2021, the tool was recognized on TIME’s “100 Best Inventions of the Year.” It was the first generative AI use case on the TIME 100, and stakeholders really loved it. Participants loved it too. Practicing with the simulation was lower-pressure than practicing with a human. Trainers loved it – they got their time back for more engaging work. And the QA team could see the difference in how new counselors actually performed. From an economics perspective, it made the entire program much more efficient while also improving outcomes. Based on this picture, we knew it was worth building beyond one organization.
For a long time, I believed the people who start companies are the ones who’ve always wanted to. I don’t believe that anymore. The way you become a founder is that you find something worth founding, and then keep improving yourself for the good of the organization. The setback I learned the most from was a deal early on at ReflexAI that I was certain would close. The customer had a problem we could solve, the relationships were strong, the ROI was clear. But, they ultimately decided to stay with their manual process. The reason was that I’d spent all my time explaining the benefits and almost none explaining how easy it would actually be to get results. So they were imagining the upside and the implementation pain at the same time, and the pain won. Now I think as much about how easy something is to adopt as I do about how much value it delivers.
Adam: How did you come up with your business idea and know it was worth pursuing? What advice do you have for others on how to come up with and test business ideas?
Sam: We didn’t come up with it as a business. We came up with it as an internal solution because we had a problem that didn’t have an off-the-shelf solution. What told us it could be a business was external demand. After the TIME 100 recognition, we started hearing repeatedly from other organizations asking what we were doing and how they could do it, too. The Trevor Project wasn’t a venue where we could support so many other organizations, so we launched ReflexAI to meet the need we were seeing and feeling.
The advice I’d give other founders is to start with customer pain. Solve a real problem your customer has today — not one they’ll have eventually, but the thing that would cause them to reach out to you right now. Then, once you’re solving actual pain, you earn the right to act more broadly to keep adding value for those same customers. For ReflexAI, we started with the core pain that manual training roleplays are time-consuming, inconsistent, and operationally complex. Once we solved that problem, now we can add even more value by accelerating training, improving outcomes, and providing insights into long-term performance that no one else can match.
Adam: What are the key steps you have taken to grow your business? What advice do you have for others on how to take their businesses to the next level?
Sam: The clearest signal for ReflexAI was customers telling other people about us. When the word-of-mouth started, we knew the product was carrying weight that didn’t depend on a sales pitch. The bar we needed to clear wasn’t impressing prospects; it was earning the kind of advocacy that gets existing customers to introduce us to new ones.
One thing that we could control was our investment in security and compliance, even before customers asked for it. We work in critical conversations that are high-stakes, regulated, often emotionally serious. We knew that as ReflexAI grew and awareness of AI increased, customers have to trust us before they can implement our tools. We pursued SOC 2, HIPAA, GDPR, HITRUST, and ISO certifications early, not because they were checkboxes, but because we would demand this level of assurance if we were in our prospects’ shoes. That investment has paid off in deal cycles, in renewals, and in the kind of accounts we win.
For other leaders thinking about the next level, my advice is two things: one cultural, one operational. The culture we’ve built at ReflexAI is what we call “very kind, very competitive.” Every person on the team is obsessed with winning and committed to being a good person and a good teammate while doing it. There’s no single personality or formula to meet this expectation, but it’s unifying. People know what we mean when we say it, and they hold themselves and each other to it.
On the operational side: lean into your environment instead of fighting it. We’re remote-first, but we don’t pretend that means everyone works whenever and wherever. We describe ourselves as “geography agnostic but time zone uncompromising.” We want the whole team online and collaborating during the same hours, no matter where they are physically. We also put real effort into remote culture — we held a talent show where people could record submissions or perform live, which is actually easier to pull off remotely. The point isn’t whether you choose remote, hybrid, or in-office. The point is to commit to your operating environment and run it deliberately.
Adam: What are your best sales and marketing tips?
Sam: The most useful tip I can give isn’t a sales tactic, it’s a metric. In service-sector work, an essential metric is time to proficiency – how long it takes a new employee to become genuinely effective at their job. It’s measurable, and shaving it down has direct business value: lower training costs, faster ramp, less attrition during the painful early phase when people don’t yet feel competent. Once we anchored many conversations there, sales cycles got shorter and more honest. Buyers know whether time to proficiency is a real problem for them or not. We don’t have to convince them, we just have to show them we can move the number.
The other thing I’d say is more about buyer behavior than seller behavior. If you’re evaluating tools in this space, the questions to ask aren’t limited to features. They’re about control and speed: will the tools give you the control you want over how they work, and can you actually deploy them quickly? Too heavy a reliance on the vendor will hold you back, even if the tool can theoretically do many things. Too many iterations before launch will kill momentum, even if the calibrated product is awesome. The companies that deserve to win in this market aren’t the ones with the flashiest demos. They’re the ones whose customers can use the product on their own terms, fast.
Adam: What are the most important trends in technology that leaders should be aware of and understand? What should they understand about them?
Sam: The most important trend for leaders to understand is that AI has moved from experimentation to expectation faster than any technology in recent memory. The internet took years for organizations to figure out. So did smartphones. Generative AI is compressing that timeline dramatically because anyone can use it at a baseline level. The bar for what employees expect their organization to be doing with AI is rising every quarter.
The harder question, and the one I think most leaders are still ducking, is not whether to use AI but how AI is actually changing the jobs and teams they’re responsible for. The consensus today is dominated by generic concepts like “human-in-the-loop,” which sound good but don’t tell you anything specific. The reality is that we don’t yet know in detail how different functions are going to evolve, and the leaders doing the best work are the ones willing to figure it out for their own context rather than borrowing a framework that fits anywhere because it commits to nothing.
For ReflexAI customers, they are interested in both higher performance in critical conversations as well as greater efficiency for their work. What makes ReflexAI compelling is that we can prove both – whether it’s greater than 50% reductions in manual training time, or over 30% improvement in priority metrics. When we can explain to a sales leader how the tools will increase their pipeline generation and pipeline conversation (to provide one example), we’re focusing on their expectations.
Adam: In your experience, what are the defining qualities of an effective leader? How can leaders and aspiring leaders take their leadership skills to the next level?
Sam: The quality I’d put first is one I haven’t seen stated this way: a built-in belief that the current version of you isn’t going to be sufficient 3-6 months from now. The leaders I’ve learned from the most operate with a real awareness that they need to keep getting better. The version of me from six months ago should be much worse at my job than the version of me today and the version of me six months from now should be more capable than today. Plateaus are often a failure of that mindset, not a failure of opportunity.
Related to that is self-awareness about how you actually think at your best. A former president of Princeton once described herself as a collaborative thinker, meaning she did her best problem-solving in conversation with others. For me, that’s true some of the time but not all of the time, and knowing the difference has mattered a lot. That ties to something I’d push back on in the conventional wisdom about leadership style. A lot of leaders adopt a fixed approach — always framing the conversation, or never sharing their opinion until everyone else has — and treat it as their signature. My experience has been the opposite. The most effective approach is the one tailored to what each situation actually needs.
A concrete example of trying to put this into practice: at our first leadership offsite, John and I picked two leaders and told them, “fill an hour with whatever you think will be most productive for the leadership team to engage on.” We didn’t specify at all. The point wasn’t to test them; it was to have the best possible session. In doing so, we also sent a clear signal that we trusted them. The offsite and working relationships were much better because of this decision.
Adam: What is your best advice on building, leading, and managing teams?
Sam: The best teams I’ve been part of are built as much (or more) through hiring than through management. In my experience, if you hire the right people, much of the downstream work of leading and managing becomes easier. So my advice is mostly about how to hire.
One question we ask candidates is to tell us about a time they had a genuine disagreement at work, not whether the headline copy should be capitalized, but a real disagreement where reasonable people came down differently. We’re listening for several things at once: what they actually disagreed about, how they approached it, and just as importantly, how they describe it now. The way someone narrates a past conflict tells you a lot about how they’ll view future disagreements.
The second thing we screen for is curiosity with receipts. Plenty of candidates will tell you they’re curious and love to learn, but fewer can show it. In the AI era, this has gotten easier to test. We want people who can describe how they’re actually using AI in their workflows, both personally (planning a road trip, sketching a meal plan) and professionally (working through a stuck draft, structuring an analysis). The detail tells you whether they’re someone who experiments occasionally or someone whose default mode is to learn and improve how they work.
Finally, hire the people you think have been underutilized somewhere else. The ones whose resume makes you wonder why no one has given them more responsibility and more resources yet. When you give those people real scope, they tend to crush it. And they tend to be loyal to the place that finally trusted them.
Adam: What are your three best tips applicable to entrepreneurs, executives, and civic leaders?
Sam: First, recognize that building a company, or any organization, is continuous. One of the strangest things about being a founder is that you have the option to change so many things on a daily basis. There are real costs to pivoting, but every day presents opportunities to make, refine, delay, and change decisions. I didn’t fully internalize the omnipresence of these options when we started ReflexAI.
Second, find a co-founder where you would trust each other to make every single decision in the organization. This is a tough tip to mention because not every leader is in a position to have a cofounder, and finding the right one is an enormous challenge. But that level of respect and trust pays off every single day. John and I have different areas of focus — I tend to spend more time on the business side, he spends more time on the product — but I would trust John to make any important decision at ReflexAI. People are surprised to hear that because the conventional view is that cofounders should focus on their domains. The relationship that’s actually mattered most for us is the opposite: full trust in both directions, regardless of who has typical ownership of any particular decision.
Third, this one is obvious, but it is always worth repeating: surround yourself with great people. For us, we were screening for investors and advisors who we would want inside our circle at the best moments and the most challenging moments. We were looking for their track records and their alignment with our approach of “very kind, very competitive.” The relationships you build with these people will shape years of your professional life and impact the company that is consuming the majority of your waking hours.
Adam: What is the single best piece of advice you have ever received?
Sam: “You have two ears and one mouth, use them in proportion.” I heard this years ago, and I continue to find new ways that it applies to leadership. First, if you have a learning mentality, you should be listening more than you’re talking. Second, you need to understand how the people around you are actually thinking, not just to steer an organization, but to help others steer it well. Third, the ratio matters in almost every part of my job. It’s true in the hiring process, and it’s true when getting advice from investors. It’s absolutely true in sales.
