Built to investigate,
not just alert.

Phalanxia was founded by a team of security researchers and AI engineers who saw firsthand how reactive most security programs had become. The industry was locked in a cycle of detect-and-respond — always one step behind attackers, always triaging instead of investigating, always shipping scores instead of verdicts.

The problem was not the technology. Behavioral analytics had matured. Large language models were becoming operationally capable. The gap was architecture: nobody had assembled those ingredients into a system that could run complete investigations autonomously — gathering evidence, tracing lateral movement, correlating identity — at the depth and speed needed to keep up with a modern adversary.

We built Phalanxia to close that gap. A multi-agent AI architecture that doesn't hand you a number and walk away, but pursues every alert through to a verdict with a full evidence trail attached. Not a replacement for your analysts — an amplification of what they can cover.

Headquartered in Seattle, Washington, with a regional office in Kigali, Rwanda, we bring together researchers, engineers, and operators who believe the future of security operations is autonomous investigation, not bigger alert queues.