Restaurant inventory AI forecasts what each dish will sell, converts that into ingredients, and suggests an order per supplier against current stock and prices. The head chef approves every order before it goes out.
All Restaurants & Food Service →
A kitchen orders ingredients, but it sells dishes. Between the two sit recipes, yields, shelf life, delivery days, minimum order quantities and the weekend's bookings. Most restaurants bridge that gap with a walk-in check and experience, which works until the chef is off, the menu changes or a supplier moves its delivery day.
Restaurant purchasing software records orders and prices well. What it rarely does is say what to order. We build that step: a forecast of covers and dish sales from the POS history, bookings, the calendar and the weather, converted through the recipes into ingredient demand, then into a suggested order for each supplier.
The forecast is back-tested on the restaurant's own history before anyone relies on it, and run alongside the chef's own orders for a few weeks. The comparison that matters is not forecast accuracy in the abstract but what each approach would have left over or run out of.
Waste is worth measuring for its own sake. In Spain, Ley 1/2025 on the prevention of food loss and waste asks food businesses to plan for it, and a record of what was ordered, used and discarded is the evidence that plan needs. How the law applies to a given business is for that business to determine.
Software that manages ordering from suppliers: catalogues, prices, purchase orders, deliveries and invoice matching. Platforms such as MarketMan, Apicbase and Restaurant365 cover it alongside inventory. Most record and track orders well; fewer forecast what should be ordered, which is the gap a forecasting layer fills.
By forecasting demand per dish and converting it into ingredient needs, so orders follow expected sales rather than habit. It also spots patterns people miss, such as an ingredient that is consistently over-ordered before a quiet day, and checks invoices against orders. The gains depend on accurate recipes and stock counts, so those come first.
It can be wired to, through supplier portals, email or EDI, but we build in an approval step. A chef knows things the data does not, such as a private event or a menu change next week, and a few seconds reviewing a suggested order is cheap. Some restaurants later let low-risk staples go automatically within agreed limits.
What we built, what stage it is at, and a link to the live thing wherever there is one.
yourOS is an operating system with an AI at its heart. You talk to it like a person, out loud or typed. It makes your calls, plans your days and works on your projects with a crew of agents, on your own private server. Our own product, built from the first commit; the waitlist is open.
Website, backend and app assets for Sonora, a free AI sound-therapy app on iOS and Android published by Cognitive AI.
Brand, website, prototype, apps on five platforms, the AI backend and the algorithm-intelligence pipeline for Podswap, a social growth engine that runs without a human in the loop.
In closed beta: Roxy, a private AI assistant with permanent memory running on infrastructure the user owns. Brand, site, product, backend and conversational design.
Live at coherence.com: brand identity generated with AI agents we directed, the motion system and website for the group behind Sonora, AboutAI, WakeUp and MudShadows. Agent work remains in build.
Live at domainbooth.com for DomainBooth, a premium domain brokerage: a new animated mark, every page redesigned and rebuilt with designed mobile layouts, and a custom CMS for its sales record, featured names, press and enquiries.
Start with a back-test: what a forecast would have ordered last quarter, against what you actually used.
30-minute technical call · no deck