One AI subscription that knows your whole family — schedules, allergies, milestones, and preferences — remembered forever, understood by everyone.
The problem
Every person in your family has their own AI. But none of them talk to each other. None of them remember what matters. You’re still the one holding it all together — and that’s exhausting.
Market opportunity
The household AI market is enormous — and completely fragmented. No one has built the shared family layer. Kin owns that category from day one.
Core pillars
Kincierge isn’t another AI chatbot. It’s your family’s personal concierge — a living, growing intelligence that knows every member, remembers everything, and executes on your behalf.
8-week build sprint
The Perplexity Billion Dollar Build competition requires a working product submitted by June 2, with finalist pitches on June 9. Here is the week-by-week build plan.
Tech stack
Every technology choice is optimized for speed to MVP — no over-engineering, no unnecessary complexity. Ship first, optimize later.
Agentic execution
Every competitor stops at reminders. Kin goes further — it acts on your behalf with full family context already loaded. This is where the product becomes truly irreplaceable.
Unit economics
Yes — AI inference at scale is expensive. Here’s exactly how Kin’s revenue model outpaces infrastructure costs as we grow, and what the bridge looks like before we get there.
Monthly cost per household
Assumes avg. 40 AI interactions/day per household. Drops significantly with caching, batching, and volume discounts at scale.
Revenue per household
73% gross margin. Every new household is immediately accretive. No physical goods, no per-unit manufacturing cost, no inventory.
Revenue vs. infrastructure at scale
| Households | Monthly Revenue | Infra Cost | Gross Profit | Margin |
|---|---|---|---|---|
| Demo (10) | $250 | ~$200 | $50 | 20% |
| Launch (100) | $2,500 | ~$800 | $1,700 | 68% |
| Growth (1,000) | $25,000 | ~$6,000 | $19,000 | 76% |
| Scale (10,000) | $250,000 | ~$45,000 | $205,000 | 82% |
| Unicorn (100,000) | $2,500,000 | ~$350,000 | $2,150,000 | 86% |
Infra costs decrease per-household at scale due to API volume discounts, prompt caching, shared model layers, and infrastructure amortization.
The unicorn moat
No competitor can replicate that data or that execution history. Every month a family stays, Kin gets smarter — about their preferences, their providers, their rhythms. It knows which dentist Nova likes, which restaurants the family returns to, which hotels worked on the last trip. That compounding context, combined with an execution layer that actually acts on your behalf, creates a product that becomes more indispensable every single day. Churn approaches zero. LTV approaches infinite. The wall gets higher with every booking made, every appointment kept, every vacation planned.
What’s next
A clear prioritized sequence from where we are today through the June 9 pitch — and beyond.
The team
Kin isn’t being built by people who’ve never run a household. It’s being built by two fathers who have lived every messy, beautiful, overwhelming moment of family life — and who know exactly what it would mean to have one AI that finally has it all figured out.