Funding & VC

Marble Raises $5.5M After Signing 150+ Restaurants

Marble raised $5.5 million on November 17, 2026, after signing 150-plus restaurants. Contrary reported that early customers cut food waste 40% and saved more than 20 labor hours weekly.

By Grace Kim

4 min read

Marble Draws $5.5M in Seed Funding After Signing 150+ Restaurants
Marble Draws $5.5M in Seed Funding After Signing 150+ RestaurantsPexels / pixabay

What's News

  • Marble announced a $5.5 million seed round on November 17, 2026, led by Contrary.
  • More than 150 restaurants were signed in under four months, according to Contrary.
  • Early customers reported 40% less food waste and more than 20 labor hours saved weekly.
  • An A&W case study reported 97% demand-forecast accuracy and 90% faster inventory counts.
  • The first A&W site went live in six days, while subsequent locations went live in hours.

By Grace Kim

Marble announced a $5.5 million seed round on November 17, 2026, after signing more than 150 restaurants in under four months, according to lead investor Contrary.

Who backed the $5.5 million seed round?

Y Combinator, Robinhood Ventures and Restive Ventures also participated, Marble said. The New York startup said the round will fund development of an AI-native operating system for restaurant back-of-house operations.

Marble identified six angel investors: Sweetgreen co-founder Jonathan Neman; Smash Kitchen and Jinx co-founder Sameer Mehta; Momos co-founder Sai Alluri; ResQ founder Kuljeev Singh; Breadcrumb founder Seth Harris; and AtoB co-founder Tushar Misra.

Marble said its platform combines inventory capture, demand forecasting, purchasing, food preparation and scheduling. The system targets the operational work behind each restaurant service, from current stock levels to supplier orders and labor plans.

The seed financing gives Marble the capital to build on its initial wave of restaurant signings and expand its product and commercial reach.

How does Remy turn restaurant data into action?

Marble divides its system into three stages: perception, reasoning and action. Employees capture inventory with voice AI or computer vision. Document extraction and third-party integrations pull invoices, deliveries, sales and staffing records into the same operating picture.

That data gives Marble a current view of stock, costs and activity across each location. Its demand forecasts incorporate sales history, weather, reservations, local events and limited-time offers to guide purchasing, food preparation and staffing.

Marble’s product documentation says the ordering workflow compares vendors and live prices with expected demand and available stock. Scheduling accounts for forecast demand, staff availability and applicable labor rules.

Marble says Remy turns those forecasts into draft supplier orders, preparation plans and schedules. The agent also reconciles invoices and flags data inaccuracies.

Marble’s product documentation says Remy “runs in the background without requiring a prompt for each task.” Operators can review its output without manually triggering each order, transfer or schedule.

Under Marble’s workflow, managers approve the resulting work. Restaurant operators retain authority over purchasing and staffing decisions, even as Remy handles more of the underlying preparation.

In short, Remy prepares operational decisions, while restaurant managers keep the authority to make them.

What have the first deployments delivered?

Contrary reported that Marble signed more than 150 restaurants in under four months. Early customers reported a 40% reduction in food waste and more than 20 labor hours saved per week.

Contrary’s account of Sophie’s Cuban in New York shows how Marble fits into a daily restaurant routine. Staff count inventory by speaking Spanish into the app, while the operator texts Remy to draft vendor orders and transfers between locations.

A separate A&W case study, published July 5, 2026, details the implementation at the first location:

  • Day one: Connect the point-of-sale system and import historical sales data.
  • Days two through four: Map the menu, recipes, vendors and par levels.
  • Days five and six: Activate Marble’s AI agent.

The A&W case study said subsequent locations went live in hours. It reported 97% accuracy for item-level demand forecasts and inventory counts completed 90% faster through computer vision and voice AI.

Those faster counts allowed A&W to take daily stock levels. Marble said the workflow can then use current inventory data rather than figures collected between infrequent manual counts.

The early deployments show two measurable gains: faster inventory work and lower waste, alongside a six-day rollout at the first A&W site.

Who is building Marble, and what comes next?

Marble said Chief Executive Officer Aakar Khanna grew up managing his family’s multi-unit restaurant business. He later covered restaurants and food and beverage as an investment banking analyst at Goldman Sachs.

Chief Technology Officer Arjun Chaliha previously led AI and machine-learning initiatives for electronic trading at Bloomberg, Marble said. Y Combinator’s profile lists Marble as founded in 2026 and a member of its Summer 2026 batch.

Marble said it is expanding its engineering, operations and growth teams. Its longer-term goal is a “self-driving back of house” that learns from each inventory count, correction and order while taking on more responsibility within controls set by operators.

The next test is whether Marble can reproduce the reported 40% waste reduction and 20-plus labor hours saved as Remy takes on more work at additional locations. Marble’s approach depends on accurate data from each kitchen and continued manager approval, so broader adoption will determine whether early results hold at scale.

Source: unite.ai

Share this article:

More from Grace Kim

Grace Kim

Show full bio

Market editor covering people, launches and day-to-day practice in entrepreneurship, startups and venture funding at Business Bearings.

2,301 articles

Related articles

« Previous articleNext article »