Nava AI, a company focused on securing autonomous financial transactions, announced that it has emerged from stealth with $8.3 million in seed funding led by Polychain Capital, with participation from Archetype, Hack VC, FalconX, Seed Club, and others. The company is building a verification layer designed to ensure that AI agents operating in decentralized finance execute transactions safely and as intended.
The platform addresses a growing risk in the emerging agent economy, where autonomous systems are increasingly executing financial actions such as trading, lending, and staking without independent oversight. Errors in prompts, incorrect parameters, or adversarial manipulation can result in unintended transactions, with limited safeguards or auditability.
Nava introduces an architecture that separates intent from execution by inserting an independent verification layer between an agent’s proposed action and the final transaction. In this model, an agent initiates a transaction, which is then reviewed by Nava’s verification system before being executed through an escrow mechanism. This process ensures that transactions are validated for intent alignment, parameter accuracy, and potential manipulation before funds move.
The system also creates an auditable record of each decision by logging encrypted transaction traces onchain. This approach is designed to improve transparency and accountability while enabling continuous learning across the network. As more transactions are processed, the platform refines its verification models, strengthening its ability to detect errors and malicious behavior across all connected agents.
Nava’s long-term vision extends beyond verification into a broader financial infrastructure layer for autonomous systems. The company plans to build additional services including payment facilitation, chargeback infrastructure, insurance mechanisms, and a native stablecoin, NavaUSD, which will incorporate built-in dispute resolution capabilities.
The founding team brings experience in blockchain infrastructure and AI safety. Vyas Krishnan was an early employee and product lead at EigenLayer, while Brianna Montgomery previously led strategy and growth at EigenLayer. Research efforts are led by Ding Zhao, who heads the Safe AI Lab at Carnegie Mellon University. The broader team includes engineers with backgrounds at EigenLayer, Consensys, Gitcoin, Lamina1, and Superfluid.
Nava is positioning itself as a foundational layer for the agent-driven financial ecosystem, aiming to enable safe participation as autonomous systems take on increasingly complex economic roles.
