Sign In Get Started

Build on OTIMO.

Deploy your first voice agent or bot in minutes, wire up telephony and 21+ connectors, and integrate deeply through webhooks, SDKs, and a full REST API. This page is a developer overview — full reference docs live on a dedicated docs subdomain.

01 // GETTING STARTED

Your first agent in 5 minutes

Web Calls need no telephony setup at all — you can test a live voice agent against your browser mic before you've configured a single phone number. Free trial credits are granted automatically so new orgs can place test calls immediately.

1
Create an agent — three ways
Agent Builder: describe your use case in plain language and the platform generates a complete starting workflow (capped at 10 generations/day/org). Blank Canvas: start from a single pre-filled Start Call node, no limits. Templates: start from a proven flow —
Appointment scheduling Restaurant reservations Lead qualification Order status lookup Healthcare scheduling Debt collections
Healthcare and collections templates ship with regulated-industry safety features already built in.
2
Build the flow
The minimal working graph is three nodes:
Start Call → Agent → End Call
"The more specific your prompts, the better the agent performs."
3
Test with a Web Call
The full pipeline — speech-to-text, LLM reasoning, text-to-speech — runs live from your browser's microphone. You see the live transcript, node transitions, and tool calls as the call happens.
4
Iterate — no redeploy
Edit a node's prompt, save, and test again immediately. Common first tweaks: refining the Agent node's tone, adding instructions to a Global node that applies across the whole call, or adding a Webhook node to send call data to your backend.
02 // VOICE AGENT & FLOW BUILDER

Design conversations visually

Both products share the same design philosophy: a node-based visual canvas with a real-engine preview, not a mock. Bot Builder ships 45+ node types across trigger, AI, condition, and action categories, with multiple named flows per bot, version history with one-click restore, and an explicit publish gate before changes go live.

Node-based visual canvas for both products
Preview runs on the real execution engine, not a simulation
Version history with one-click restore
Explicit publish gate before changes go live
03 // TELEPHONY

Connecting Twilio — real steps

Seven native telephony providers in total — Twilio, Vonage, Plivo, Telnyx, Cloudonix, Vobiz, and self-hosted Asterisk ARI. Twilio is the fastest to connect:

Sign up at Twilio (free trial credits available)
Copy your Account SID + Auth Token from the Twilio dashboard
In Telephony, "Add Configuration" → select Twilio → enter credentials
Add your Twilio number, set country, select an inbound workflow
On trial Twilio accounts, verify the destination number before testing
Gotcha: forgetting to mark a configuration "set as default for outbound calls" produces a 400: Telephony provider not configured error when you trigger a test call via the API.
04 // CONNECTORS

One shared connector ecosystem

All 21 connectors are available to both Voice AI and Bot Builder — there's no separate integration list per product. Sync leads, tickets, and conversation data with the systems your team already runs on.

HubSpot, Salesforce, Zendesk, Slack, Zapier
Dynamics 365, Shopify, and 15 more
Same 21 connectors across voice and bot workflows
05 // TOOLS & KNOWLEDGE BASE

Keeping agents grounded: two different mechanisms

Both must be explicitly attached to an agent node — creating a tool or knowledge base alone isn't enough to make an agent use it. Your system prompt should tell the agent which mechanism answers which kind of question: "don't leave it to guesswork."

🔧 Tools
Dynamic data or actions — anything live, anything that changes, or anything that needs to happen. Documented type: External HTTP API (e.g. an unauthenticated GET request to check order status or look up availability).
"Anything live, data that changes, or an action that needs to happen, like checking an order or looking up availability."
📚 Knowledge Base
Static content that doesn't change from call to call — a return policy, an FAQ. Two retrieval modes: Full Document (small files, agent sees the whole thing, no embeddings model required) and Chunked Search (large documents, agent retrieves only the relevant section, requires an embeddings model).
"Anything written down, like your return policy or an FAQ, that doesn't change from call to call."
06 // ENTERPRISE / SELF-HOSTED DEPLOYMENT

Run it on your own infrastructure

Most teams don't need this section — requesting access or accepting an invite on the hosted platform is enough. Self-hosting is the option for organizations that want their own infrastructure, deployed via Docker Compose.

api container ui container PostgreSQL Redis MinIO
Local Development
Runs on your own machine. No public URL, no telephony — good for building and testing flow logic.
Remote Server
Internet-accessible with HTTPS via sslip.io certificates — no custom domain required to get a working, publicly reachable deployment.
Custom Domain
A branded URL (e.g. voice.yourcompany.com) for production deployments you want under your own domain.
Note: browsers require a secure context for microphone access — HTTPS in production, or localhost / 127.0.0.1 over HTTP for local development. Ongoing self-hosted operations include auth configuration, scaling for call volume, version updates, and security/compliance upkeep.
07 // API REFERENCE

Every endpoint, fully documented

Programmatic access to agents, bots, campaigns, runs, and telephony/channel configuration — including triggering calls from your own backend and receiving call data via webhooks. Full reference documentation covering every endpoint lives on our dedicated docs site.

GET /v1/agents
POST /v1/agents/{id}/runs
POST /v1/calls/trigger
GET /v1/campaigns
POST /v1/telephony/configs
GET /v1/bots/{id}/flows
// Ready to build

Start with your own keys,
or let us run it for you.

Full API reference and step-by-step guides are available once you're set up. Talk to us to get access.