Brian Kuan
Building independent, tamper-evident records of what AI agents do to provide a verifiable audit trail for security and GRC teams.
At a Glance
- AI Agent Vendors
- Enterprise GRC Teams
- Security compliance professionals
AI Tools by Brian Kuan
(1)halo-record
AI Agent Audit Trail Library
Discussions
No discussions yet
Be the first to start a discussion about Brian Kuan
Latest News
Release of halo-record 0.2.4 on PyPI
Halo-record hits 50+ stars on GitHub as AI agent security interest grows
Major release 0.2.0: Introduced policy engine and OWASP/AIUC-1 starter packs for AI agent compliance
Halo AI selected for Y Combinator 2026 batch
Products & Services
Open, tamper-evident runtime records for AI agents using the Halo Runtime Record format. Includes a reference implementation for recording, verifying, and witnessing agent actions.
Market Position
Halo AI differentiates by focusing on tamper-evident 'evidence' for compliance rather than just debugging telemetry. It aims to be the 'SOC 2 for AI agents' equivalent in terms of record-keeping.
Leadership
Founders
Brian Kuan
Previously at Vanta (Security & Compliance) and Samsara (IoT/Operations). Employee #4 at Spot AI. Experience in Product Marketing and GTM Strategy. Y Combinator alumnus (2025/2026).
Executive Team
Brian Kuan
Founder
Ex-Vanta and Samsara. Expert in security, compliance, and GTM strategy for startups.
Founding Story
Security reviews already ask AI questions next to the SOC 2 checklist, but today written assurances still pass. Brian Kuan started Halo AI on the bet that verifiable evidence will soon be required for AI agent compliance.
Business Model
Revenue Model
Open source / Free for implementation (Apache-2.0). Potential future commercial offerings around witnessing and reporting infrastructure.
Pricing Tiers
Apache-2.0 license, full access to recorder and verifier.
Target Markets
- AI Agent Vendors
- Enterprise GRC Teams
- Security compliance professionals
- Security and compliance audits for AI agent vendors
- Providing evidence of agent behavior to customer GRC teams
- Ensuring transparency and trust in autonomous AI systems
- Self-verifying runtime reports for AI applications
- Noma Security