Skip to main content
Build and deploy your first AI agent without installing anything.

​
Before you begin

​
Ways to build on the platform

The platform supports three paths depending on your team’s skills and workflow:

​
Setup guide

Set up your first multi-agent system in four phases:

​
Phase 1: Access studio

Sign up and log in Access the platform login page and create your account. After verifying your email, you land in Studio - the browser-based IDE where you build, test, and manage your agents. Create your first project Click New Project from the Studio dashboard. Give it a name like “My First Agents” and select a workspace. Projects organize your agents, supervisors, tools, and knowledge sources in one place.

​
Phase 2: Build your first agent

Inside your project, click New Agent. Open the ABL editor and paste this definition:
This definition creates an agent that:
  • Uses an LLM to understand customer questions.
  • Searches a knowledge base for answers.
  • Responds with sourced information.
  • Has clear boundaries on what it can and can’t do.
Click Save to validate your definition. Studio parses ABL in real time and flags syntax issues inline.

​
Phase 3: Test your agent

Open the Test panel on the right side of Studio and send a message:
Your agent processes the message, searches for relevant knowledge, and responds. The trace viewer below the chat shows the full execution - LLM calls, tool invocations, and reasoning steps. Send a few more messages to see the agent handle different questions, maintain context across turns, and respect its defined limitations.

​
Phase 4: Add a supervisor

Create a new Supervisor in your project and paste this definition:
The supervisor evaluates each incoming message and routes it to the right agent, passing conversation context along. Test it the same way - open the Test panel and send messages that should route to different agents.

​
What have you built

In a few minutes, you created:
  • An agent that understands natural language, retrieves knowledge, and enforces boundaries.
  • A supervisor that routes messages to the right specialist.
  • Observable traces for every execution step, visible right in Studio.

​
Platform access and allowlist of IPs

The Agent Platform is available in multiple regions. The Platform sends requests to a fixed set of egress IP addresses. If your systems use IP restrictions, add the region-specific egress IPs below to your allowlist to permit Platform traffic.
Related article
  • Your first agent tutorial - Build a complete agent with tools, knowledge, and guardrails.
  • MCP server - Use platform’s MCP server to build, evaluate, optimize, debug, and analyze agents.
Check out the Template Gallery in Studio for ready-made agent definitions across industries - airlines, retail, banking, telecom, travel, and more.