Fabraix

Description

Verify AI agents through adversarial testing services that identify unsafe behaviors and weaknesses before deployment, based on adversarial verification research.

Related

• MirrorsBuilds tools that route contextual guidance into AI agent conversations, helping agent teams use trace observations, preview matched…
• LangChainBuilds open source frameworks and platforms for teams developing agents, providing tools to control complex agent systems and observe AI…
• TensorTestBuilds private evaluations, RL environments, and training data for AI agents on creative jobs, helping AI labs test unreleased systems and…
• ImbueBuilds AI tools and an open agent ecosystem for people and small teams to understand, modify, verify, and control software agents serving…
• CodagReduces coding-agent tool output and costs by filtering large command results for Claude Code and Codex through a local proxy before agents…
• One RobotBuilds robot simulation environments and world models that help robotics teams test policies, find edge cases, and generate synthetic…
• GoodfireBuilds interpretability agents and infrastructure that help AI teams understand, monitor, debug, and align model behavior across language,…
• DeepMarkEmbeds machine-readable identity into AI-generated media and live synthetic speech, helping teams mark, detect, and verify generated…
• VerdexVerdex builds an AI insurance adjuster using satellite imagery and weather models to verify claims quickly for insurers, especially those…
• Alkera AIBuilds an agentic data platform that connects existing data tools, reads knowledge-base context, maintains business definitions, and checks…
• Earendil RoboticsBuilds systems that intercept inbound threats, share battlefield awareness across operators, provide thermal sight in darkness, and support…
• MemorableBuilds graph-based procedural memory for AI agents, capturing session traces, converting successful runs into reusable workflows, and…