Serendipidate - AI Simulated Dating

Inspiration

First dates are often terrible. Around 80-90% of first dates don’t lead to a second, often due to simple compatibility issues that could have been identified early on. What if dates could be simulated before they happen, ensuring users meet people they’re more likely to connect with from the start, saving time, reducing frustration, and increasing the likelihood of meaningful, lasting connections. What if dating felt more real? People waste time on shallow connections and mismatched dates.

What it does

Serendipidate simulates dates with AI. Answer a few questions about yourself and we will create an AI agent of you that will go on 1000s of simulated with other users. A walk on Venice beach or a hike in the mountains, Serendipidate can simulate it all!

People often don't know the best first date spot either and Serendipidate can suggest this from all the simulated dates as well!

How we built it

We used react native to build the mobile app, Kindo for AI orchestration and Firebase for the backend. We built an AI agent profile of each user using Kindo. Kindo was really key to help us choose the relevant model and quickly switch for all our use cases, without building loads of different API calls. We used many different models in many different contexts and Kindo was key to the success.

Challenges we ran into

It is hard to get AI to output things you want. Additionally large amounts of testing meant we frequently ran into rate limits especially when trying to simulate loads of dates at scale.

Accomplishments that we're proud of

Building the app, synthetically making data and making AI agents go on virtual dates!

What we learned

Using a variety of models is key for this, faster models rapidly improved certain tasks

What's next for Serendipidate - AI Simulated Dating

As users engage with Serendipidate, they build a rich data persona that goes beyond dating. Our long-term vision is to allow this persona to unlock new experiences across other verticals, like job matching, event discovery, and social networking. This creates an ecosystem where users seamlessly transition between apps, using the same personalized data profile. Each new vertical unlocks additional revenue opportunities through tiered subscriptions and cross-app integration, driving deeper engagement and lifetime value.