Skild AI, the Pittsburgh robotics startup that rocketed to unicorn status after a blockbuster fundraising run, has locked in a partnership that will put its robot-control software on Nvidia Blackwell-powered production lines run with Foxconn. The deal drops a hometown player into the hardware supply chain that turns out high-end GPUs and will test whether general-purpose robotics software can actually scale on messy, real-world factory floors. For local tech workers and researchers, it signals that Pittsburgh’s AI scene is edging out of the lab and onto the line.
As reported by the Pittsburgh Business Times, Skild will plug its "Skild Brain" software into robots working on Nvidia Blackwell GPU production lines through a collaboration that involves Foxconn. According to the Business Times, the arrangement grew out of Nvidia and Foxconn’s broader push to stand up Blackwell-based AI factory production environments.
Skild has spent the past year in full sprint. The company closed an approximately $1.4 billion Series C in January that included NVentures, NVIDIA’s venture arm, according to a Business Wire release. Skild’s own blog describes the "Skild Brain" as a single foundation model meant to run across many different robot bodies and improve through real-world deployments, per Skild AI.
Why Nvidia And Foxconn Matter
NVIDIA has been pitching its Blackwell GPUs and Omniverse and Isaac software stack as the backbone of so-called "AI factories," and announced a collaboration with Foxconn to deliver Blackwell infrastructure to industry users and researchers. NVIDIA says those factory designs are meant to speed up simulation, training, and deployment of robot fleets on a large scale.
What Skild Brings
Foxconn’s materials describe using RTX PRO servers, Omniverse digital twins, and Isaac robotics tools to simulate and operate automated production lines, exactly the kind of environment where a general-purpose robotics model can be trained and stress-tested. Per Foxconn, those systems are being rolled into pilot plants, and Skild’s technology is designed to run inside the same simulated-to-real training loops. Skild AI says fleet deployments generate the data the company needs to scale the Skild Brain.
Pittsburgh's Place In The Supply Chain
Carnegie Mellon University and a wave of local startups have factored into Nvidia’s regional outreach, with the chipmaker helping launch an AI Tech Community in Pittsburgh and pointing to area researchers as likely beneficiaries of factory-scale tools. Carnegie Mellon University has highlighted Skild as one of the local teams working on general-purpose robotics models.
What To Watch Next
NVIDIA’s recent financial release name-checked Skild among a roster of manufacturing and robotics partners as the company pushes Blackwell into production use cases, underscoring the commercial stakes of the tie-up, per NVIDIA. The Pittsburgh Business Times reports the companies have not yet laid out a detailed public timeline for full rollouts.
For Skild, the real prize is scale. Running on the same factory floors that build Blackwell GPUs would generate the production data the company says it needs to sharpen its foundation model. Company statements and investor materials describe deployments and a growing data flywheel as central to Skild’s growth plan, according to the Business Wire release.
