Destro AI has emerged from stealth with an $8 million seed round and a warehouse pilot that puts coordination ahead of inventing another robot body. Its software directs cart-moving machines and gives human workers instructions within the same logistics workflow. The company and a customer argue that this is more useful than automating only one movement between two points.
The round was led by Base10 Partners and Bonfire Ventures, with CoFound Partners also participating, according to TechCrunch. The investment follows a pilot at one Yusen Logistics facility in the Pacific Northwest. Destro is expanding that deployment and preparing a second pilot, but the article offers no independent measurements of cost savings, throughput or safety outcomes.
A warehouse problem sets the design
Yusen Logistics already uses fixed equipment such as conveyors and sorters in its US operation and is testing other machines, including trailer-unloading robots and autonomous floor scrubbers. Richard Brunelle, its director of automation for the American logistics group, oversees roughly 30 US facilities. His immediate problem was not finding a novel-looking robot. It was coordinating cross-docking: unloading goods from one truck and sorting them into mixed loads for onward delivery.
When Brunelle met Destro founder Manthan Pawar, the startup was focused on picking and packing. Brunelle asked whether its approach could be adapted to cross-dock work. Destro accepted the change, and the two sides began with a pilot at one Yusen site. This customer-led shift is central to the story: the system was shaped around an existing warehouse sequence rather than a demonstration task designed for a robot.
The pilot used three cart-moving robots made by Miva Robotics. They were operated by Destro’s Vision system, which the report says uses open-weight vision-language-action models to interpret images and instructions for movement. Human workers unloaded goods into carts; robots located full loads and carried them to their destinations. Destro’s Mothership system directed the people, carts and truck-related steps as one operation, according to the company and Brunelle.
Coordination rather than a new machine
Brunelle told TechCrunch that the pilot made work less labour-intensive and replaced paper with a more systematic process. That is a customer account, not a published before-and-after study. He said two other robotics suppliers he considered could not fit the workflow: one moved carts between points without managing loading and unloading, while another offered fleet management but left orchestration to a person.
Destro’s pitch is therefore narrower than general-purpose warehouse autonomy. It connects existing robot capability to the human decisions and hand-offs required by a specific logistics task. Pawar, who has a robotics master’s degree and about eight years of US supply-chain and robotics experience, argues that knowing the customer’s process matters more to the company than building hardware itself. His claim that Destro is on a path to cash-flow positivity by year-end remains a projection.
That distinction also explains why the startup can use machines from another manufacturer and open-weight models. Its proposed value lies in the system around them: deciding where a cart should go and how that movement fits with work by people and trucks. TechCrunch describes the arrangement but does not test it against an independent benchmark or quantify how often interventions are needed.
Expansion, funding and limits
Yusen is expanding the initial pilot to a full deployment of 26 robots. Destro is also launching a 17-robot pilot at a Yusen facility in Southern California. Pawar wants to repeat the cross-dock approach across many other warehouses, a commercial ambition that has not yet been demonstrated at that scale.
The company may face different problems where work requires fine manipulation. TechCrunch notes that generic robot bodies and open models have not proved themselves for those tasks. It is also unclear whether makers of more dexterous hardware or foundation models would make their technology available to Destro rather than entering the workflow market themselves.
Brunelle says automation must solve an operational problem rather than be installed for its novelty. The current evidence shows one pilot being expanded and another beginning, backed by a seed round. Whether Destro’s coordination layer delivers measurable gains across different sites and more difficult tasks remains an open question.
