How a torch-wielding robot from Ohio could solve America’s welder shortage by 2030

America’s advanced manufacturing sector is at a crossroads. While venture capital pours billions into cutting-edge industries like data centers, defense technology, and robotics, a fundamental challenge remains: a severe shortage of skilled welders.

The American Welding Society estimates that the U.S. will require more than 320,000 new welding professionals by 2030—meaning the country must hire approximately 80,000 new welders annually to meet demand. Without enough trained hands to melt, fuse, and repair metal, the future of advanced manufacturing could stall.

From Failed Custom Vehicle Startup to Robotic Welding Pioneer

Path Robotics, a Columbus, Ohio-based company, believes the solution lies in augmenting the workforce with torch-wielding robots. Founded in 2018 by brothers Andy and Alex Lonsberry, the company emerged from a personal struggle: their attempt to launch a custom vehicle business with their father.

“It was four grueling years seeing what it was truly like to try to stand up a U.S. manufacturing company, and it was extremely tough and ultimately ended with failure,” said Andy Lonsberry, CEO of Path Robotics. “It gave us a deeply rooted passion and understanding for how difficult it truly is to be a small to medium-sized U.S. manufacturing firm, and that experience has stuck with us.”

Introducing Rove: The Mobile Welding Robot for Shipyards and Beyond

Path Robotics is now expanding into the shipbuilding industry with Rove, a new welding robot mounted on the back of a Boston Dynamics quadruped. Unlike traditional fixed robotic arms, Rove is designed for mobility and adaptability—ideal for environments like shipyards, where human welders often need to navigate uneven terrain, climb scaffolds, or reach awkward angles.

The robot combines Boston Dynamics’ agile robotic platform with Path’s proprietary Obsidian AI system. Once deployed, Rove’s robotic arm extends, a laser scans the welding site, and its torch ignites to close seams with precision. The system relies on extensive training and real-world testing to ensure flawless performance.

How Obsidian AI Enables Adaptive Welding

Training an AI system to weld is no small feat. “There’s no room for error when the wrong connection can ruin a project,” Lonsberry explains. Unlike traditional assembly-line robots that repeat the same motion millions of times, Path’s technology is designed to execute a million different moves, each once—making it far more versatile.

“Robots have been involved in manufacturing for 50 years,” Lonsberry notes. “But that’s mainly been in automotive, where they’re doing the same motion over