Traversal raises $120 million at a $1 billion valuation

Traversal announced on 3 November 2026 that it had raised $120 million in Series B funding at a $1 billion valuation. The startup builds superintelligence (SI) software for site reliability engineering, the work of keeping live software systems running and investigating failures.

Sequoia Capital and Battery Ventures also participated, Traversal said. SiliconANGLE said Amex Ventures joined the round as well. Kleiner Perkins and Sequoia had invested in Traversal before, the company said.

Investigating failures in live systems

Traversal says its software connects operational data—including logs, metrics and traces—with code and deployment history, then uses causal machine learning, which studies cause and effect rather than relying on patterns alone, to test possible causes of a failure. SiliconANGLE describes the software as recommending a fix for a human engineer to accept.

Co-founder and CEO Anish Agarwal has a background in causal machine learning. Traversal says nearly two years of research and engineering led to the introduction of its Production World Model and Causal Search Engine in March 2026.

Agarwal cites ORCA-bench, a test Traversal developed with researchers from Cornell Tech and Columbia University, as evidence of the difficulty of the task, not independent validation of its product. He says advanced models achieved roughly 25% accuracy on medium-difficulty reliability tasks and often produced made-up answers more than 40% of the time.

Growth and hiring plans

Traversal says its revenue grew more than ninefold over the last year, while its monthly active user count rose 20 times. The company names PepsiCo, DigitalOcean, Nubank and Harvey among its customers.

Kleiner Perkins investor Aditya Naganath predicts demand will grow as SI speeds up software delivery and makes live systems more complex to manage. He says Traversal can help engineering teams understand their systems and resolve incidents faster.

With the new funding, Traversal plans to hire across research, engineering and commercial teams. It aims eventually to build systems that detect problems, identify their causes, take corrective action, check the results and prevent future incidents without human intervention. Agarwal says the company aims to help enterprises keep pace with AI-enabled development teams.