News
Marble raises $5.5M to build the self-driving operating system for restaurants.
Marble now powers the back of house for some of the world's leading restaurant and hospitality brands, and we're excited to share we've raised $5.5M led by Contrary, with participation from Y Combinator, Robinhood Ventures, Restive Ventures, and a host of strategic angels.

As reported by Restaurant Business, we've raised a $5.5M seed round led by Contrary, with participation from Y Combinator, Robinhood Ventures, and Restive Ventures, alongside strategic angel investors*: Jonathan Neman (co-founder of Sweetgreen); Sameer Mehta (co-founder of Smash Kitchen and Jinx); Sai Alluri (co-founder of Momos); Kuljeev Singh (founder of ResQ); Seth Harris (founder of Breadcrumb); Tushar Misra (co-founder of AtoB); and individuals with current or former affiliations at OpenAI, DoorDash, Momofuku, and the a16z Scout Fund.
We're building toward a world where AI can run the operations behind physical businesses. That requires systems that can perceive what's happening, reason about what comes next, and carry decisions through to action. We're starting with the industry we know best.
Why restaurants
You don't need to have read the National Restaurant Association's 2026 State of the Restaurant Industry report to know operators are under pressure. But the numbers are damning: 42% of operators reported their restaurants were unprofitable in 2025, and 82% reported higher food costs than the year before.
I spent over a decade helping run my family's multi-unit QSR franchise. In doing so, I realized that restaurant owner-operators are presented with a practical paradox: the work that draws people to hospitality is orthogonal to the work that determines whether they can sustain a business in it. Profitability often comes down to a team's proficiency in supply chain management, labor allocation, demand forecasting, and working capital management.
And even so, while modern technology solutions help restaurants bring more customers through the door, the decisions that ultimately determine whether those visits are profitable still rely on a patchwork of outdated point solutions.
Getting these decisions right has enormous upside. Across millions of restaurants and hospitality venues worldwide, even small improvements in purchasing, waste, and labor planning can create substantial value. For operators, that means more room to invest in the guest experience, open additional locations, or weather a challenging month.
Bringing AI into the physical world
Much of the work AI excels at today remains in the digital world, where documents, code, transactions, and structured records already capture the context the system needs. A restaurant generates critical information every minute, but much of it never reaches a database. That includes what's left in the walk-in, what came off the truck, how vendor prices shifted, and what was prepped, wasted, or substituted during service.
An agent making decisions about that operation needs an accurate understanding of the ground truth. Building that understanding means connecting what happens in the kitchen with sales, inventory, invoices, staffing, and the signals that shape demand. The system must then continuously update that understanding as conditions change.
Capturing this context is the central technical challenge in automating physical businesses. Software can record a purchase or report a variance based on entered data, but building a real-time representation of the operation requires context that legacy software doesn't capture. Bringing perception, reasoning, and action into one system is what makes this such a consequential problem in applied AI.
What Marble does
Marble connects three capabilities under one roof:
- Perceive. Capture ground truth through AI-powered inventory tracking, document extraction, and dozens of third-party integrations. A stock take, a delivery receipt, and an invoice need to resolve into a consistent understanding of what a location has and what it costs.
- Reason. Forecast demand from sales history, weather, reservations, local events, LTOs, and more. Carry that forecast into decisions about what to purchase, how much to prep, and how to staff each location, accounting for the stock and constraints already in place.
- Act. Unlike traditional chatbots, our agent, Remy, works autonomously in the background without requiring explicit prompting. Drawing on ground truth data and an understanding of broader business objectives, it drafts orders, builds prep plans, reconciles invoices, flags data inaccuracies, and prepares schedules for managers to review and approve.
Our vision
Our vision is a self-driving back of house that continuously learns from the operation it supports. As its understanding improves, our goal is for Marble to take on more operational responsibility, within the controls operators set. That is how we see AI becoming part of the infrastructure of a physical business.
Our investors
This round was led by Contrary, with participation from Y Combinator, Robinhood Ventures, and Restive Ventures.
“Over the course of my career, I've been an investor in Toast and Square and looked at dozens of restaurant-focused point solutions. None of them solve the biggest black hole for restaurant profitability that is the rat's nest of problems in the back-of-house. Within weeks of starting, Aakar and Arjun had already ripped out a dozen existing solutions. Restaurants don't do that, unless there's proven value. This is the problem the Marble team has lived and is rapidly deploying this across multi-unit operators at a shocking pace.”
Our angels include Jonathan Neman (co-founder of Sweetgreen); Sameer Mehta (co-founder of Smash Kitchen and Jinx); Sai Alluri (co-founder of Momos); Kuljeev Singh (founder of ResQ); Seth Harris (founder of Breadcrumb); Tushar Misra (co-founder of AtoB); and individuals with current or former affiliations at OpenAI, DoorDash, Momofuku, and the a16z Scout Fund.
What's next
We're a small team with outsized impact, and we're growing across engineering, operations, and growth. If you're passionate about your work, deeply curious, and excited to build systems whose decisions reach beyond a screen, we'd love to meet you.
And if you run a restaurant, cafe, or bar and these challenges sound familiar, we'd love to hear from you.
More to come soon.
*Angel investors participated in their personal capacities. Affiliations are provided for identification only and do not imply investment or endorsement by their employers.