The Kindo Platform is an AI-native system for automating enterprise technical operations across SecOps, DevOps, and ITOps functions. It centralizes the management and deployment of autonomous agents that ingest alerts, logs, vulnerability scans, network data, and threat feeds, normalizing and enriching this information with asset, identity, and intelligence context. The platform utilizes Deep Hat, a proprietary large language model purpose-trained for infrastructure, security, and incident response tasks, while remaining model-agnostic and compatible with over 26 LLMs from providers including OpenAI, Anthropic, Google, and Meta. Agents interpret natural language, read human-written runbooks, and execute multi-step actions including API calls, script execution, and tool integration via the Model Context Protocol (MCP). The system supports five agent types—Chatbot, Manual Workflow, Scheduled, Triggered, and Dashboard—with memory patterns including run-scoped memory, per-user per-agent persistent sandboxes, ephemeral sandboxes, a Knowledge Store for organization-wide static reference data, and external system integrations for cross-workflow state sharing. Agents coordinate on tasks such as vulnerability management, IAM enforcement, incident response, infrastructure automation, ticketing, compliance auditing, continuous control monitoring, evidence collection, and red-team simulation.
The platform can be deployed as SaaS, self-managed in cloud environments, or on-premises with no internet requirement, including installation inside customer Kubernetes clusters using the kindo-cli tool. Self-managed deployments require customer-provided infrastructure including Kubernetes, PostgreSQL, Redis, RabbitMQ, S3-compatible storage, DNS/TLS, and AI model endpoints. Security and governance controls include single sign-on (SSO), role-based access control (RBAC) with additive group permissions and organization-wide defaults, audit logging of all interactions and configuration changes, data loss prevention (DLP) rules using Presidio for PII and credential detection, built-in encryption, a dedicated secrets manager, human-in-the-loop approvals for AI-generated actions, and policy-driven guardrails. The platform integrates with existing tools such as Jira, CrowdStrike, Slack, GitHub, ServiceNow, SIEMs, source control, CI/CD pipelines, and cloud services through MCP and native connectors, provides secure sandbox environments for isolated execution, offers REST APIs for inference and agent execution, and supports a Git repository pattern for version management of agent configurations. Compliance certifications include SOC 2 Type II with support for ISO 27001, GDPR, and NIST CSF frameworks.
