Traversal, a New York startup whose AI agents figure out why company software breaks, has raised a $120 million Series B at a $1 billion valuation, the company announced Tuesday, per The Next Web.

The pitch is elegantly self-referential: as companies lean harder on AI to write code, more software ships faster — and more of it breaks in ways human engineers struggle to trace. Traversal sells AI agents that do the detective work, hunting down the root cause of failures across sprawling production systems. It is site reliability engineering, automated.

The round makes Traversal one of the newest entrants to the unicorn club, and the valuation says a lot about where investors think the AI economy's pressure points are forming. The money flooding into code-generation tools has a second-order consequence: a growing mountain of software that someone — or something — has to keep running. Startups positioned at that cleanup layer are increasingly attractive precisely because they win regardless of which coding assistant dominates.

A $1 billion price tag at Series B leaves little room for a modest outcome. Traversal is now capitalized to chase the enterprise observability and incident-response market at full speed — and to prove that debugging, the least glamorous job in software, might be one of AI's most valuable.