Designing AI-Native Features for a Consumer App
Designing AI-Native Features for a Consumer App
At Schmooze I lead product strategy for a set of AI-native features in dating: a Personal Matchmaker, a People Finder, a Dating Coach and AI Companions. Working on them has changed how I think about what makes an AI feature worth shipping in a consumer product.
Start from a real moment of need
AI features are easy to build as demos and hard to build as habits. The ones that work start from a moment where a user genuinely needs help: not knowing who to talk to, not knowing what to say, or wanting a second opinion. If the feature does not map to a moment like that, it becomes something people try once and forget.
Conversation is the interface
Conversational AI changes the shape of a product. Instead of screens and buttons, the user is talking to something. That makes tone, pacing and honesty part of the product design. A matchmaker or coach that sounds generic breaks trust quickly. Personalisation is what makes the conversation feel worth having.
Measure it like any other feature
It is tempting to judge AI features by how impressive they look. I prefer to hold them to the same standard as the rest of the app: do they improve activation, engagement or monetization? That keeps the team honest and stops the roadmap from filling up with features that look good in a launch post but do not move users.
Safety is part of the experience
In a product about relationships, people are sharing things that matter to them. AI features have to be designed with that in mind, with clear limits on what they do and say. Trust is slow to earn and quick to lose, and it applies to the AI as much as to the rest of the product.
What the cohort added
Building AI products in the Curious PM cohort gave me more structure for testing and evaluating these features before they reach users, which matters more as they take on bigger roles in the app.

Jyotsna skipped presentations and built real AI products.
Jyotsna Akurathi was part of the August 2026 cohort at Curious PM, alongside 24 other talented participants.
