Destro AI Is Building the Shared Mind for Warehouse Robots

Destro AI is building a shared intelligence layer for warehouses at the point when buying another robot no longer solves the operating problem. A warehouse can have capable autonomous mobile robots, forklifts, robotic arms, software systems, and experienced people while still lacking one reliable answer to the question that runs the floor: what should happen next?

Founded in 2025 and based in Brooklyn, Destro is led by founder and CEO Manthan Pawar and co-founder and CTO Soorya Narayanan. The company is developing MothershipOS and VisionOS as complementary systems for coordinating work outside the robot and handling perception and manipulation on the robot. The goal is not to make customers loyal to one machine. It is to make the operation coherent across many machines.

That argument now has an $8M Seed round behind it. Base10 Partners and Bonfire Ventures co-led the financing, with CoFound Partners participating. Destro plans to expand enterprise deployments, advance both products, and grow its engineering and research teams.

About Destro AI

Destro starts from a useful distinction. Robot autonomy describes what one machine can do. Warehouse intelligence describes how the building decides which machine or person should do which task, under which constraint, at which moment.

MothershipOS is Destro's off-robot coordination layer. The company says it observes robots, people, inventory, orders, and operating conditions, then plans and assigns work as priorities change. VisionOS is the on-robot intelligence layer, using perception and manipulation models to identify products and translate higher-level instructions into physical action.

The two systems divide the problem cleanly. One tries to understand the floor. The other tries to understand the object in front of the machine. Destro's commercial thesis is that both forms of intelligence must share context if automation is going to survive outside a demo.

Why Warehouse Coordination Is Becoming the Product

Warehouse robotics has spent years improving hardware. The result is a growing menu of autonomous forklifts, mobile robots, arms, mobile manipulators, and specialized systems. That variety gives operators more choice, but it also creates more interfaces, control systems, and vendor dependencies.

Destro is betting that workflow logic should sit above that hardware turnover. A customer should be able to change a robot without rebuilding the business rules that determine how freight, inventory, labor, and service commitments move through the facility.

That is a meaningful idea, but it is not automatic. A robot-agnostic control layer still needs dependable integrations, accurate state data, graceful failure behavior, and clear boundaries between machine decisions and human authority. The software becomes more valuable as it sees more of the operation, and more consequential when it is wrong.

Production Evidence at Yusen Logistics

The strongest public deployment record comes from Yusen Logistics. In August 2026, the companies announced a collaboration around transload operations, where inbound freight, staging, dock activity, labor availability, and outbound commitments change continuously.

The initial deployment coordinates employees and autonomous mobile robots handling cart movements. Yusen said it expects to evaluate pallet movement and AI-powered workflow verification as the work develops. Rick Brunelle, Yusen Logistics' director of automation, described the platform as a way to give teams more intelligence and control while supporting service expectations.

Destro also says it generated commercial revenue and entered production with large third-party logistics providers less than a year after founding. Those are company-reported claims. Public sources do not provide audited revenue, a customer count, site totals, contract values, or independently measured performance results.

That restraint does not erase the signal. A named logistics operator using the system in an exception-heavy workflow is more useful evidence than a polished robot video. It places Destro inside the environment where integration failures, network problems, equipment faults, and shifting priorities become operating costs.

The Founders and the Floor-First Culture

Manthan Pawar built his perspective through warehouse robotics and deployment work, including experience connected to HAI Robotics and Dexterity. Bonfire Ventures says customers consistently understood his value proposition and trusted his grasp of real deployment pain. Soorya Narayanan brings human-robot collaboration expertise that Bonfire calls foundational to the product vision.

The company describes its culture as floor-first. Engineers and researchers are expected to work alongside operators in production rather than treating the warehouse as the place software visits after the lab is finished. That posture matters because physical AI inherits every messy constraint that digital demos can avoid: intermittent connectivity, unpredictable objects, safety boundaries, changing service levels, and machinery that does not care what the roadmap promised.

Destro's careers page points toward work across software, field deployment, operations, business development, enterprise sales, and robotics support. The site's visible role counter is inconsistent with the individual role pages, so an exact opening count would be misleading. The durable hiring signal is the kind of responsibility Destro is trying to add: production reliability, customer integration, fleet operations, and enterprise deployment.

Capital, Investors, and Market Timing

Base10 Partners and Bonfire Ventures co-led Destro's Seed round, while CoFound Partners participated. Unshackled Ventures also identifies Destro as a portfolio company, though the amount of that earlier backing is not public. Destro has not disclosed its valuation, ownership terms, or complete capital history.

The investor logic is clearer than the cap table. Base10 invests in automation across the real economy. Bonfire focuses on early B2B software and framed its Destro investment around the gap between a successful robotics demonstration and a successful deployment. Both firms are underwriting the idea that physical AI creates value only when it improves a live operation.

The $8M gives Destro more chances to prove that idea across customers and facilities. It does not prove repeatability, safety, uptime, or return on investment. Those questions now become the company's real product roadmap.

What Destro AI Signals for Physical AI

Physical AI is often described as a race to build more capable machines. Destro represents a different layer of the market: the software that decides how machines, people, inventory, and enterprise systems should coordinate after the hardware reaches the floor.

That layer could become durable because warehouses rarely standardize around one vendor forever. Hardware changes, supply chains shift, regulations move, and new automation arrives one workflow at a time. The intelligence that preserves operating context across those changes can become more strategic than any single machine.

Destro is worth watching because its thesis is simple enough to test. If MothershipOS and VisionOS can help mixed operations handle exceptions, change hardware, and improve business outcomes without sacrificing control, the company will have built more than robot software. It will have built a way for the warehouse to keep its mind while everything underneath it changes.