Senior Product Manager, AI

This hiring round has closed. The challenge below is still available for practice — completing it now is a learning experience, not an application.

About the role

You'd lead one of our two AI product lines: the **AI feedback product**, which teachers use to give students faster and more substantive written feedback on their work. It's our most-used surface and our hardest to get right. Teachers don't want AI to grade for them. They want AI to help them be the version of themselves they don't have time to be at 11pm with 90 essays left. You'd own the roadmap, work directly with two engineers and a designer, and partner with our learning science lead on what "good feedback" actually looks like in a 9th grade ELA classroom vs. an AP Chem lab vs. a community college composition section. ## What you'd actually do - Sit with teachers in their classrooms and at their kitchen tables on Sundays. Watch them work. Find the moments where the product makes them feel either more or less like the teacher they want to be. - Make hard calls about what AI should do vs. what we leave to the human. This is the central design question of our entire category, and there's no playbook. - Write specs engineers want to build from. Clear on the user, the problem, and the why. Light on the what. - Run experiments. We ship weekly and instrument heavily. You'd be comfortable looking at funnels, but more interested in whether the thing actually made the teacher feel respected. - Help shape how we talk about AI in education externally. Teachers are skeptical for good reasons. Schools are skeptical for different good reasons. You'd help us be honest with both.

What we're looking for

We care about how you think, not where you went to school or what your last title was. The strongest signal for us is the ability to take a messy problem space (say, "what does it mean for AI to help a teacher without replacing their judgment?") and structure it well enough that a team can build against it. Concretely, we'd want to see evidence that you can: - Hold a strong product opinion and update it quickly when the user or the data tells you something different. - Get to the actual problem underneath what people say they want. - Write clearly enough that a stranger could follow your reasoning. - Work close to the model. You don't have to be technical, but you have to be curious about *why* a particular AI behavior is happening, and willing to learn enough to have a real conversation with engineers about it. - Care about education. Not the industry. The thing. Teachers, classrooms, what learning actually feels like when it's working. If you've built or shipped something (a product, a research project, a side thing, an internal tool at a previous job) and can walk us through the decisions you made and what you'd do differently now, that's more interesting to us than a résumé full of impressive logos. ## Helpful, but not required - PM experience at an AI-native company - Experience in K-12 or higher ed, as a teacher, instructional designer, or someone who built tools for them - Background in learning sciences, cognitive science, linguistics, or a related field - Having taught anything to anyone, formally or not - Comfort with SQL, Python notebooks, or evals tooling We don't require a CS degree, an MBA, a specific number of years, or having held the title "Product Manager" before. If you've done the thinking and the work, we'd rather see that than a credential.