Co-Founder
Arcade
Arcade provides the MCP runtime for deploying AI agents that securely execute actions across various systems. It enables agents to connect to APIs, data sources, and services using granular permissions and agent-optimized tools. The platform allows developers to build and deploy functional AI applications that move beyond simple conversation to perform real-world tasks.
Updated 17 days ago
Funding
Raised to date
$12MRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
Funding rounds are not available yet.
Founders
Product
Problem
AI agents are typically confined to conversational interfaces and lack reliable, secure mechanisms to perform actions on external services, making it difficult for developers to build production‑grade, autonomous workflows. Integrating with OAuth‑protected APIs and managing token lifecycles also adds significant engineering overhead. Consequently, organizations cannot efficiently deploy AI‑driven automation across the diverse SaaS tools they already use.
Solution
Arcade delivers a Multi‑Cloud Platform (MCP) runtime that bridges large language models and real‑world services. The platform grants AI agents granular, revocable permissions to any OAuth‑enabled or secret‑protected API, eliminating token‑management headaches. Developers can start instantly with a library of battle‑tested connectors (e.g., Gmail, Slack, Google Calendar) or extend functionality via a custom tool SDK that generates agent‑optimized wrappers. Centralized governance lets teams audit, version, and benchmark LLM‑tool interactions to ensure consistent performance. The runtime can be hosted in the cloud, within a VPC, or on‑premises, providing flexible deployment for any security posture. An integrated marketplace lets creators publish and monetize reusable tools, accelerating the build‑once‑use‑many model for AI‑enabled automation.
Target Audience
The primary customers are software engineers and AI product teams building autonomous agents, as well as enterprise IT groups that need secure, auditable AI‑driven workflows across their SaaS stack.
Features
- Agent‑auth layer that handles OAuth flows, token refresh, and secret storage, providing per‑user, per‑tool access controls.
- Pre‑built connectors for over 50 popular SaaS products (Gmail, Slack, Google Workspace, Salesforce, etc.) with zero‑code integration.
- Custom Tool SDK (Python & JavaScript) to define typed tool schemas, automatic request signing, and error handling for any REST or GraphQL endpoint.
- Tool evaluation suite that runs automated benchmark suites to measure LLM‑tool success rates, latency, and cost, enabling data‑driven tool selection.
- Centralized governance dashboard for permission audits, usage analytics, and policy enforcement across all deployed agents.
- Flexible deployment options: managed Arcade Cloud, self‑hosted MCP server in Kubernetes, or on‑prem VM, all with TLS‑encrypted communication.
- Marketplace/Registry for publishing, versioning, and monetizing reusable agentic tools across the developer ecosystem.