Kindo

Kindo is an enterprise AI platform built for security operations, centered around its proprietary cybersecurity AI model called Deep Hat. The platform enables autonomous, agentic workflows for security teams, allowing AI agents to independently plan, execute, and self-correct across a wide range of security tasks. Core capabilities: - Deep Hat: A proprietary, uncensored red team cybersecurity AI model trained specifically for the cybersecurity domain, including penetration testing, APT simulation, and vulnerability discovery. Deep Hat v2 uses a Mixture of Experts (MoE) architecture with a 256,000-token context window. - Agentic Framework: An autonomous AI loop that selects tools, executes actions, and self-corrects on errors by feeding error messages back into the model as new context. A two-step tool filtration process prevents context overload. - Human-in-the-Loop Controls: Administrators can review and approve AI-planned actions before execution, preventing unintended automated changes to sensitive systems. - Self-Managed Deployment: The entire Kindo stack, including the Deep Hat model, can be deployed within a customer's own environment, ensuring sensitive data never leaves the organization. - Administrative Controls: Administrators can define which tools are accessible to AI agents and block access to sensitive or dangerous systems. - Security Use Cases: Incident response acceleration (reducing APT investigation from days to under an hour), automated red and blue team exercises, continuous risk assessments, penetration testing (DPAT v2), and automated security compliance documentation generation. - SIEM/SOAR Replacement: Uses dynamic runtime AI to replace static, manually built SOAR workflows. Users describe objectives in natural language and the AI autonomously navigates tools to resolve alerts. - Non-Security Use Cases: Non-technical staff can build custom AI agents for tasks such as marketing newsletters and HR training modules.