Plurai opens agent testing platform after raising $13 million in seed funding
The AMW Read
The launch and seed round add a provider to agent testing, while its reported performance gains lack disclosed methods and named customers.
Plurai opens agent testing platform after raising $13 million in seed funding
Plurai launched its AI agent testing platform and announced a $13 million seed round led by Team8. U&I, Velocity, Mercer and more than 20 angel investors also participated. Previously available to selected partners and customers, the platform is now open to agent developers, including through a free plan. Plurai did not disclose how it will use the funding.
The launch adds a testing provider to the AI agent market, where developers must assess how systems behave across varied interactions. Plurai says its platform creates digital twins of work environments and simulates thousands of realistic and malicious users. It also generates proposed fixes and post-trains a small language model for each agent. The platform supports agents used through Claude Code, Codex, GitHub Copilot and Salesforce Agentforce.
For builders, the practical question is whether simulated interactions reveal failures that matter in their own production workflows. Plurai says design partners cut development time from months to hours, reduced errors reaching users by 82%, and lowered the cost of stable agent operations by 97%. Those are company-reported results; the article does not provide the underlying methods or identify the customers. Buyers should assess the platform against their own tasks and costs before assuming similar gains.
