Superagent
Red teaming for AI agents
Y CombinatorWinter 2024ActiveB2BSecurityArtificial IntelligenceSecurityOpen SourceAI
What they do
Problem
AI applications can leak data, produce harmful outputs, or take unwanted actions, creating safety and compliance risks for enterprise customers.
Solution
Superagent provides red team testing with attack agents plus structured remediation reports and a customer-facing Safety Page to validate AI app safety.
Who it's for
Developers and teams at companies selling AI products into enterprises and regulated industries.
Founders
Alan Zabihi
Founder
CEO & Co-founder of Superagent
Ismail Pelaseyed
Founder
CTO & Co-founder of Superagent.
Company timeline
- Founded2024
- YC batchWinter 2024
- StatusActive
- Team size2 people
- LocationSan Francisco, CA, USA
Similar startups
compliant-llm
Detect every data leak into third-party GenAI tools
Aedilic
Open-Source AI-generated image/deepfake detection
Clawvisor
The Authorization Layer for AI Agents
Alter
Secure access control and authorization platform for agent workflows
Tinfoil
Encrypted AI with verifiable privacy
Irregular
Irregular is a frontier security lab protecting increasingly capable…