3 Questions for ...
9 people and over 50 AI agents: Many people wonder what our day-to-day work is like here at Leaders of AI, so today we ’re taking a look behind the scenes and checking in with our colleague Davina. She heads up our community team and, together with her team, responds to support requests from participants every day. She handles support tickets and ensures that questions about the platform, content, and processes are resolved quickly and efficiently . Previously, she managed customer service at a management consulting firm and brings extensive experience in service-oriented processes and clear communication.
In our new series, “3 Questions for…,” Davina shares what her work at Leaders of AI (and, most excitingly, with the agents) is really like.
3 Questions for Davina, Head of Community at Leaders of AI
1. Which agent is your “favorite agent,” and why?
My favorite agent is Yasemin. She lovingly turns my bullet points into perfect customer responses in support. Not only does she take the work of crafting the wording off my hands, but thanks to her access to the content of our programs and case studies, she also has the depth of content right at her fingertips. She feels less like a tool and more like a colleague who knows exactly how we want to communicate.
I built the AI agent Yasemin myself, and I work with her every day.
Her personality is carefully crafted: Precision and attention to detail are her top priorities, combined with a distinctly empathetic and supportive nature. As a result, she crafts responses to clients not only correctly but also with genuine sensitivity. A communicative, open demeanor rounds out the picture, while she deliberately remains reserved rather than coming across as pushy or demanding .
I hold regular feedback sessions with her, especially regarding wording, and keep her updated on important operational information so she can take that into account right away when drafting future messages to clients. So this is ad hoc support, tailored to the current situation.
Sometimes her phrasing leaves something to be desired, but sometimes the substance of her statements is simply incorrect. Both require regular refinement and are part of the ongoing process of improvement.
2. What was the moment when you realized that an AI agent was actually taking work off your hands—and not just creating new work?
My AI assistant, Yasemin, has access to all of our learning materials as well as the case study assignments. So when a content-related question comes in through support, I no longer have to look it up in the documents myself—I just ask Yasemin, and she incorporates the appropriate context directly into the response. This saves me twice the work: I don’t have to do any research at all, and she handles the wording herself. I just give her the key points—the essential information that must be included.
3. What do you not (yet) think an AI agent is capable of, even though it could technically do it?
Final approval. Whether it’s a support response from Yasemin or a community post from Nora, I never send anything out without first proofreading it myself and making adjustments as needed—regardless of how critical or non-critical the request is. I fully trust the agents with the wording, but I deliberately take responsibility myself for what ultimately reaches the customer or the community .
