Restauranty

Inspiration

Restaurant no-shows create a weird lose-lose situation:

restaurants lose revenue from empty tables
staff have to adjust last minute
diners who wanted the table never get a chance
reservation resale/scalping makes the problem worse

We wanted to build a more responsible system where restaurants stay in control and empty tables can be recovered through verified waitlists.

What it does

Restauranty helps restaurants prevent no-shows and refill empty tables with verified diners.

Restaurant owners can:

claim/register their restaurant
manage reservations
configure cancellation and recovery policies
detect risky reservations
approve table recovery actions

Diners can:

confirm they are coming
release a reservation responsibly
join waitlists for tables that may open up

The key idea is that this is not reservation scalping. Restauranty is a restaurant-controlled recovery system.

How we built it

We built Restauranty as a full-stack web app with a production-style architecture:

MongoDB Atlas for restaurants, reservations, policies, waitlists, audit logs, and agent events
Google Places API for real restaurant search and restaurant claiming
Auth0 for user authentication and role-based flows
Fetch.ai / Agentverse-inspired agents for risk scoring, policy checks, waitlist matching, and recovery recommendations
Figma Make / Cloud Design to prototype the UI and guide the final dashboard design

The agent workflow is split into smaller roles: Risk Agent, Policy Agent, Waitlist Agent, Fairness Agent, Restaurant Agent, and Recovery Agent.

Challenges we ran into

The biggest challenge was making sure the product felt ethical and restaurant-friendly. We had to avoid building anything that looked like a resale marketplace.

Other challenges:

moving from seeded demo data to real Google Places restaurant data
setting up persistent database flows with MongoDB
handling auth and different user roles
making the frontend feel like one polished product instead of separate demo screens
designing agent logic that is useful but still deterministic and auditable

Accomplishments that we're proud of

We’re proud that Restauranty has a real product flow:

restaurant search and claiming
restaurant dashboard
reservation management
no-show risk scoring
waitlist recovery
policy enforcement
audit logs
sponsor-track integrations

We’re also proud that the AI/agent layer is not just a chatbot. Each agent has a clear job, and restaurants still make the final decision.

What we learned

We learned that for agentic products, the workflow matters as much as the AI. The most useful part is not “predicting the future,” but creating earlier intervention points before a table becomes a no-show.

We also learned that trust is central in restaurant tech:

restaurants need control
diners need transparency
waitlists need fairness
recovery actions need auditability

What's next for Restauranty

Next, we would want to:

integrate with reservation/POS systems like Resy, OpenTable, Tock, Toast, or Square
add real SMS confirmation and waitlist flows with Twilio
strengthen restaurant ownership verification
register the recovery agent on Agentverse
improve diner verification with World ID
build better analytics for recovered revenue and no-show trends

Long term, Restauranty could become a restaurant operations layer for reducing no-shows and improving table utilization without enabling scalping.