EESI AI can be deployed using Docker in two main configurations. Choose the option that best fits your needs:
  • Option 1: For local development and testing on your own machine
  • Option 2: For remote server deployment with HTTPS (using IP address). If you also have a custom domain, you can first deploy EESI stack on your server using steps in this document and then proceed to the Custom Domain section.
Using Claude Code or Codex? Install the official EESI setup skill and let your agent drive either deployment below — it orients to your OS, picks local vs remote, runs EESI’s own setup scripts, and verifies the result with a built-in health check.
Start a new session, then ask it to “set up EESI” (or run /eesi-setup in Claude Code). More at eesi-ai/eesi-plugins.

​
Option 1: Local Docker Deployment

Watch the video tutorial below for a step-by-step walkthrough of setting up EESI AI locally with Docker.
For local development and testing, you can run EESI AI directly on your machine using Docker with a small startup script.

​
Quick Start

Download the compose file and starter script, then confirm the prompt to start EESI AI:
This setup:
  • Downloads the latest docker-compose.yaml
  • Creates REDIS_PASSWORD and MinIO credentials in .env if they do not already exist
  • Creates POSTGRES_PASSWORD for brand-new .env files and syncs retained local Postgres volumes to that value before startup
  • Prompts before running Docker Compose
  • Starts all required services including PostgreSQL, Redis, MinIO, API, and UI
  • Pulls the latest images automatically

​
Access the Application

Once running (first startup takes 2-3 minutes), open:
You can disable telemetry by setting ENABLE_TELEMETRY=false before running ./start_docker.sh or .\start_docker.ps1.

​
Troubleshooting WebRTC Connectivity

The Quick Start above relies on direct peer-to-peer WebRTC between your browser and the API container. On most single-machine setups this works without any extra configuration. If you hit any of the following, you likely need a TURN server in the path:
  • The call connects but no audio flows in either direction
  • The browser console reports iceConnectionState: failed
  • You are testing from a browser on your smartphone or another device on your LAN against the laptop running Docker
  • A VPN, corporate firewall, or strict NAT sits between the browser and Docker
For these cases, use the alternate local setup script which configures a coturn TURN server alongside the rest of the stack:
The script will prompt you for:
  • Whether to enable coturn (answer y)
  • The host browsers should use to reach TURN (press Enter for 127.0.0.1; use your LAN IP if testing from another device on the same network)
  • A shared secret for the TURN server (press Enter to generate a random one)
It creates docker-compose.yaml, a .env file with JWT and TURN credentials, and the small helper bundle that eesi-init uses to render coturn config at startup. Start the stack with the local-turn profile so coturn comes up alongside the other services:
The application is still available at http://localhost:3010.
To verify that media is actually traversing TURN (and not silently falling back to a direct path), set FORCE_TURN_RELAY=true in .env and restart the API. The browser will then only use relay ICE candidates — if TURN is misconfigured or unreachable, the call fails cleanly with ICE failed in the console instead of appearing to work. Set it back to false once you have verified connectivity.

​
Option 2: Remote Server Deployment

Watch the video tutorial below for a step-by-step walkthrough of deploying EESI AI to a remote server.
Deploy EESI AI on a remote server to make it accessible from anywhere. If your server has a public IP, setup obtains a free, trusted HTTPS certificate automatically via sslip.io and Let’s Encrypt — no domain name or DNS configuration required, and no browser warning. On a private/reserved IP it falls back to a self-signed certificate. We need to serve the application over HTTPS, since modern browsers only allow microphone permissions for websites being served over HTTPS.
We highly recommend you set up the platform on a fresh server, so that there are less chances of confliciting dependencies, and ports from other applications.

​
Prerequisites

  • A server with Docker and Docker Compose installed. It should have minimum of 8 GB RAM and 4 vCPUs.
  • Root (sudo) access on the server — the setup script must be run as root and exits early otherwise.
  • Public IP address for your server. You can also access the server using a local IP address in your VPC as long as its reachable from your browser.
  • TCP Ports 80, 443, 3478, 5349 and UDP Ports 3478, 5349 and 49152:49200 reachable from Internet (Port 80 and 443 to access the UI and rest of the ports for WebRTC Signaling)
IMPORTANT: Please double check the ports on your firewall. Please refer to your server hosting provider’s documentaion on how you can open these ports in the firewall of the server.

​
Quick Setup

Run the automated setup script that will configure everything for you:
Run with sudo — the script must run as root (it provisions Docker, binds ports 80/443, and installs a Let’s Encrypt certificate with a system renewal hook) and exits early if it isn’t. On a public IP it obtains and auto-renews a trusted certificate; on a private IP it falls back to a self-signed one, since Let’s Encrypt can’t validate a private address. Optional environment-variable overrides: CERT_MODE=self-signed forces a self-signed cert, ACME_DOMAIN_SUFFIX=nip.io switches to the nip.io pool if sslip.io is rate-limited, and LETSENCRYPT_EMAIL=you@example.com sets the address for expiry notices.
The script will prompt you for:
  • Your server’s public IP address
  • A password for the TURN server (optional, press Enter for default)
  • Deployment mode — press Enter for prebuilt (pulls official images, the recommended default) or pick build to compile images from source. Use build when you maintain a fork or want to deploy local customizations — see Building from source below.
  • Number of FastAPI workers (uvicorn processes nginx will load-balance) — press Enter for the default of 4, or pick a value that suits your server’s CPU count. See Scaling for sizing guidance and how to change this on a running stack.
It will automatically:
  • Get the source — docker-compose.yaml only (prebuilt mode), or clone the full repo (build mode)
  • Download the validated remote deployment helper bundle
  • Obtain a free trusted Let’s Encrypt certificate via sslip.io and configure auto-renewal (public IP), or generate a self-signed certificate (private IP / no root)
  • For a public IP, also start the stack and print your ready-to-use https://…sslip.io URL
  • Create an environment file with TURN server configuration
  • Validate the runtime config that eesi-init will render from .env
  • Write a docker-compose.override.yaml with build directives (build mode only)

​
Start the Application

Please ensure that Docker Compose is installed on your machine before proceeding further. You can check whether its installed by running docker compose version command. If its not installed, please install it by following your server provider documentation.
After the setup script completes, start EESI. For a public-IP install the trusted-certificate flow already started the stack and printed your https://…sslip.io URL — you can skip to Access Your Application. For a self-signed install, start it yourself (the script prints the exact command at the end — it differs slightly between modes):
First boot in build mode takes several minutes — Docker has to build both the API and UI images before the stack comes up.

​
Access Your Application

Your application will be available at the URL the setup script printed at the end:
  • Public IP (trusted certificate): https://<your-ip-with-dashes>.sslip.io — for example https://203-0-113-10.sslip.io. No browser warning.
  • Self-signed (private IP): https://YOUR_SERVER_IP
With a self-signed certificate your browser shows a security warning you can safely accept. With the sslip.io Let’s Encrypt certificate there is no warning.
You should be able to create and test a voice agent now.

​
Configuration Notes

  • The remote deployment includes an nginx reverse proxy for HTTPS termination
  • File downloads (transcripts, recordings) are automatically routed through nginx
  • WebSocket connections for real-time features are properly proxied
  • Public-IP installs get a trusted Let’s Encrypt certificate (via sslip.io) that renews automatically; private-IP installs use a self-signed certificate (browser warning)
  • The TURN server (coturn) is configured for WebRTC NAT traversal
  • For production deployments with proper SSL and domain names, see the Custom Domain documentation

​
Files Created

The setup script creates the following files in the eesi/ directory:

​
Building from source

If you maintain a fork or want to ship local code changes without waiting for an official image release, pick build when the setup script prompts for deployment mode. In this mode the script:
  1. Clones the repo into eesi/ (default: eesi-ai/eesi@main; the script will ask for an owner/name and branch so a fork can point at its own). The deployment then runs from eesi/infra/docker/.
  2. If you run the script from inside an existing checkout’s infra/docker/ directory, offers to use that checkout instead of cloning.
  3. Generates a docker-compose.override.yaml next to docker-compose.yaml that swaps the image: directives for local build: directives. Docker Compose auto-loads the override file — you don’t need to pass any -f flags.

​
Updating after a code change

Whenever the codebase changes — a git pull, a local edit, or a submodule bump — rebuild the api and ui images and recreate the running containers so they pick up the new code. TL;DR (rebuilds both services and restarts in one shot):
--build rebuilds any service whose build: context has changed (both api and ui are wired through the docker-compose.override.yaml written by setup_remote.sh), and up -d recreates the affected containers so they run off the new image. Step-by-step — use this when you’re pulling fresh code from a remote, or want to be explicit:
If you update pipecat submodule, you must do git submodule update --init --recursive, or the Docker build step will not pick up pipecat changes
Rebuild a single service — faster when you only changed one side:
Force a clean rebuild — needed when a base image changed, a pip install was added, or you suspect a stale layer:
Verify the containers were actually recreated — check the CREATED column shows a recent timestamp, not minutes/hours ago:
To revert to pulling official images, delete docker-compose.override.yaml and start the stack with --pull always as in the prebuilt flow.
Build mode is for fork maintainers and self-hosters who want to deploy customized images — not for active development on the code itself. For day-to-day contributor work (fast reload, IDE/LSP integration), use the contributor setup instead.

​
Next Steps

Now that you are able to create and test a voice agent, you can setup a custom domain and setup SSL using letsencrypt. Checkout Custom Domain for instructions on how to do that.