Relari (YC W24)
Testing, Evaluation and Synthetic Data for AI Agents
🛡️ AgentReady threat assessment
MAESTRO 7-layer threat model + OWASP AIVSS risk score for Relari (YC W24), derived from its capabilities.
These scores are auto-generated from public information (the agent's own listing, docs, and repository) using the canonical OWASP AIVSS formula and the MAESTRO framework — an estimate for guidance, not a penetration test, audit, or certification. See the scoring methodology. Are you the vendor? Factual corrections are free.
Overview
Relari is a platform that helps AI teams simulate, test, and validate complex Generative AI applications throughout the development lifecycle. It offers modular evaluation, synthetic data generation, and performance monitoring tools to improve the reliability and efficiency of AI systems, particularly for mission-critical use cases. Define test cases for agents in natural language using Agent Contracts. Expand test cases by 100x with Synthetic Data Generation. Pinpoint issues and effortless improve your Agentic application.
Key features
- Modular evaluation framework with 30+ open-source metrics
- Synthetic test-set generators
- Online monitoring tools
- Custom evaluators trained on user feedback
- Continuous evaluation for AI pipelines
Use cases
- Pinpointing root causes of problems in LLM applications
- Simulating user behavior for AI system testing
- Accelerating AI development with synthetic data
- Stress testing GenAI applications before deployment
- Improving reliability of complex AI systems in finance, enterprise search, and compliance