JeliciaRoss

I came into my career genuinely curious and not particularly attached to a fixed destination. Consulting was a good fit for that. Every engagement was a new problem, a new domain, a different set of constraints to figure out. What I was really building in those years was a way of working: get into something quickly, build something real, learn from what breaks, and keep going.

When data science started becoming a serious discipline in the industry, I wanted in. I had the instincts for it and the drive to develop the skills. Getting there took longer than it should have, and not because of any gap in my ability. That experience clarified something important about what I wanted to spend my career doing and who I wanted to build it for.

At some point in every data project, I noticed that the numbers were never really the thing people needed. What they needed was to understand what the numbers meant about them: their decisions, their culture, what they were reinforcing without realizing it. That became the work I cared about most.

I also do improv and sketch comedy, and it turns out the skill is the same. You are not trying to be funny. You are trying to understand what the room needs and build the conditions for it to land. A dashboard that no one acts on has the same problem as a joke with no setup. The content might be right, but the communication failed.

So I focused on people analytics not as a reporting function, but as a way to show organizations an honest picture of themselves: what they had built, what they were enabling, and what needed to change. That lens changed how I approached everything.

When I joined Pinterest, the team I was stepping into did not really exist yet as a distinct function. I built it: the structure, the culture, the processes, the tooling. I also spent 11 months as Interim Director across two teams after my manager left, which meant doing both jobs at once while continuing to ship. That was hard, and it also confirmed what I already suspected about how I work best: I do my clearest thinking when the problem is real and the stakes are not hypothetical.

The AI experimentation I have done at Pinterest, with Claude, MCPs, and HRIS integrations, has been less about what the tools can do and more about understanding where they fall short, where people get frustrated with them, and what a better version of the pairing looks like. That is the problem I find most interesting right now.

I believe the next generation of people analytics tooling should be designed with the same care and intentionality we expect from any product that affects how people experience work. That is the standard I hold myself to, and the one I want to keep building toward.