StarDust
Find your match in the fandoms you love
Be among the first to get access
The problem with mainstream dating
The billion-dollar dating app industry has a dirty secret: it is optimised for engagement, not for actual connection. Swipe fatigue is real. Profile surfaces are shallow. And if your identity revolves around fandoms—anime, gaming, Dungeons & Dragons, sci-fi, fantasy literature—you are invisible on every major platform.
There is nowhere that understands the difference between a casual fan and someone whose entire social life is built around a particular universe. There is nowhere to signal that you're a "read the books three times" person versus a "watched the movie once" person. And there is certainly nowhere that lets you create or join dedicated community spaces built around shared fandoms.
So people cobble together Discord servers, Reddit threads, and Twitter DMs—none of which are optimised for meeting romantic partners. The community exists. The infrastructure doesn't.
A platform built for your passions
StarDust is an AI-powered dating and community platform designed from the ground up for niche communities. The core insight is simple: shared identity is the strongest predictor of compatibility—and fandom is one of the richest identity signals available.
The platform has two interlocking layers. The dating layer uses AI to surface compatibility based on deep interest graphs—not just "likes anime" but which shows, which characters, which eras, which adjacent fandoms. Matches are meaningful because the signal is specific.
The community layer lets users create and join fandom Spaces—structured communities around specific universes. Spaces are not just chat rooms; they're structured around events, shared media, discussion prompts, and meetup tools. You can find romantic partners inside a Space you already love, or join Spaces because someone you matched with is deeply involved in them.
Cross-pollination between the two layers is where the magic happens: a match leads to a shared Space, a Space conversation leads to a date. The social graph is coherent in a way that no current platform achieves.
Built for scale and safety
StarDust is built on a modern, scalable stack chosen for speed of iteration and long-term reliability.
Flutter powers the mobile client—a single codebase for iOS and Android that allows rapid feature development without sacrificing native performance. The UI is opinionated: dark-first, content-rich, and built around the aesthetic sensibilities of the communities we serve.
Encore.ts handles the backend, providing type-safe API development with built-in observability, infrastructure provisioning, and cloud-agnostic deployment. Backend services are structured as microservices: auth, profiles, matching, spaces, messaging, and AI inference are all separate, independently deployable units.
PostgreSQL on GCP serves as the primary data store. The matching engine runs on top of pgvector for embedding-based similarity search—fandom interests are encoded as high-dimensional vectors and match candidates are retrieved by nearest-neighbour lookup. This makes AI-driven matching fast and explainable.
Safety and trust infrastructure is built in from day one: AI-assisted content moderation, verified profile flows, rate limiting, and community governance tools for Space moderators.
Where we are
StarDust is currently in private beta with a closed group of early testers. We are validating core flows: profile creation, fandom onboarding, Spaces creation, and matching quality.
The waitlist is open. F&F (Friends & Family) launch is the next milestone—a controlled rollout to a few hundred users to stress-test the social dynamics before wider release.
Post F&F, we're targeting an App Store submission. The beta feedback so far has been strong on the community layer; matching quality is the area receiving the most active iteration.
Be among the first
StarDust is built for the communities that have been overlooked by mainstream dating apps. If you're passionate about anime, gaming, sci-fi, fantasy, or any other fandom—this is for you.