Hinge

Redesigning Dating Around People

The Unknown

The gap wasn’t visible in feature metrics. It showed up in behavior.

When this work began, dating at Hinge was treated as a solo activity.

Profiles were created alone. Photos were selected alone. Prompts were written alone. The system assumed that authenticity was an individual responsibility. At the same time, broader cultural signals were shifting.

Outside the product, I was seeing a clear pattern:

  • Rising discourse around loneliness and dating fatigue

  • Cultural conversation framing dating as isolating, not connective

  • A growing gap between how people actually shape identity (with others) and how dating apps expected them to do it (alone)

This gap wasn’t visible in feature metrics. It showed up in behavior.

People were already pulling others in informally:

  • Sending screenshots of profiles to friends

  • Asking group chats for prompt advice

  • Crowd-sourcing photos through text threads and shared albums

None of this signal was captured by the product. None of it informed the system. And none of it fed into Hinge’s emerging AI strategy. The risk wasn’t that features weren’t working, it was that the product model was misaligned with how dating actually happens culturally.

The Mission

Expand how the organization understood authenticity and signal, and to do it in a way that strengthened the AI-first strategy rather than distracting from it.

I originated and took ownership of a new hypothesis:

If AI is meant to understand people more deeply, it needs access to how people are understood socially, not just individually.

This idea was not on any roadmap. It was not being explored by Labs. Leadership was skeptical due to historically failed attempts at other companies. My goal was not to propose social features.

My goal was to expand how the organization understood authenticity and signal, and to do it in a way that strengthened the AI-first strategy rather than distracting from it.

I treated this as a small, contained exploration, knowing that if the insight was real, the implications would be large.

The Launch

Shifting the Strategic Frame

The core strategic move was deciding where support should enter the system.

Rather than intervening downstream after matching—where other apps had failed, I focused exploration upstream, while someone is becoming visible. The insight was structural: people already involve friends in dating, but the product ignores it.

Exploring Upstream Social Support

I explored bringing people’s circles into identity formation through concepts such as:

  • Profile feedback from friends instead of screenshots

  • Friends elevating how someone shows up

  • Shared photo libraries friends could contribute to

  • Lightweight comments or short clips from friends on profiles

This surfaced a hidden cultural metric the product had never captured: does this person have friends?

Framing This as an AI Strategy Win

Internally, I positioned this work carefully:

  • As a contained exploration, not a roadmap proposal

  • As aligned with Hinge’s AI-first strategy

  • As a way to generate more authentic upstream signal

The framing was explicit. Social input would give AI a better understanding of the dater, strengthening systems like Prompt Feedback without adding pressure or gimmicks.

To estimate potential impact, I worked with leadership to use proxy signals such as screenshot rates to gauge unmet demand. The goal was not feature validation, but strategic signal.

Fast, High-Fidelity Prototyping

I built and tested concepts using vibe coding tools to move quickly while maintaining enough fidelity for real reactions. This demonstrated internally that AI and social concepts can be tested rapidly, but only if fidelity is high early. Speed and rigor were not opposites.

Expanding Research Itself

I introduced and led Hinge’s first-ever friends pair study. This shifted research from individuals to collaboration and observed identity creation as a social act.

This built directly on the broader expansion of Labs’ scope, continuing the shift from pure insights toward execution-ready product exploration.

The Resolution

This work reframed how social support fits into dating products.

Measurable Outcome

The strongest signal came after the sessions, not during them. Research participants independently updated their profiles based on friend input. Those updates drove measurable lifts in actual dates, the hardest metric to move at Hinge.

This direction was not on any leader’s radar. It was not an explicit Labs initiative. Even Labs initially did not expect or believe in it. I used influence to make the case, reduce perceived risk, and connect it to existing strategy by keeping it small, safe, and reversible.

The result was legitimacy. Teams believed in it. The exploration gained traction without requiring a big bet upfront.

Strategic Impact

This became a category-level unlock inside Hinge. It shifted the framing from downstream optimization to upstream understanding and became a named pillar of Hinge’s 2026 strategy.

It did not just propose social features. It proved a new way to learn about daters, improve AI inputs, and align the product model with culture, ahead of the trend rather than reacting to it.

This work reframed social support as a win for the core AI strategy. By bringing in more authentic upstream signal, we improved our understanding of the dater, creating stronger inputs for AI systems like Prompt Feedback without adding pressure or gimmicky mechanics.