Riven AIDocs

Quick Start

Get up and running with Riven AI in under 5 minutes.

Prerequisites

Before you begin, make sure you have the following:

  • Node.js 22+ — Required for the CLI and TypeScript SDKs.
  • Docker 24+ — Used for building and running container images locally.
  • kubectl 1.28+ — For interacting with your Kubernetes cluster.

You will also need a Riven AI account. Sign up at riven-ai.dev if you do not have one yet.

Step 1: Install the CLI

Install the Riven CLI using the install script:

Terminal
bash
curl -fsSL https://docs.riven-ai.dev/api/install | bash

Verify it installed correctly:

Terminal
bash
riven --version
Expected output
text
@riven-private/riven-cli v0.2.10

Step 2: Authenticate

Log in to the Riven platform. This opens your browser for OAuth-based sign-in:

Terminal
bash
riven auth login
Expected output
text
Opening browser for authentication...
✓ Logged in as [email protected] (org: riven-ai)
Credentials saved to ~/.config/riven/credentials.json

Step 3: Create a Project

Scaffold a new project using the init command:

Terminal
bash
riven init my-project
cd my-project
Expected output
text
Creating project my-project...
✓ riven.config.ts
✓ src/index.ts
✓ Dockerfile
✓ helm/
✓ .gitignore
 
Project my-project created successfully.

Use the --template flag to start from a pre-built template. For example, riven init my-project --template ai-service scaffolds a project with model serving already configured.

The generated project includes:

  • riven.config.ts — Project configuration file.
  • src/ — Your application source code.
  • Dockerfile — Container build definition.
  • helm/ — Helm chart for Kubernetes deployment.

Step 4: Deploy

Deploy your project to your Riven cluster:

Terminal
bash
riven deploy
Expected output
text
Building container image...
✓ Image built: my-project:1.0.0-a1b2c3d
Pushing to registry...
✓ Pushed to <account-id>.dkr.ecr.us-east-1.amazonaws.com/my-project:1.0.0-a1b2c3d
Deploying to cluster production (namespace: default)...
✓ Deployment complete
 
Service URL: https://my-project.riven-ai.dev

Troubleshooting

riven: command not found

The CLI binary is not on your PATH. If you used the install script, restart your terminal or run:

Terminal
bash
source ~/.bashrc  # or ~/.zshrc

If you installed via npm, ensure your global npm bin directory is in your PATH:

Terminal
bash
export PATH="$(npm config get prefix)/bin:$PATH"

Error: Not authenticated

You need to log in before deploying. Run riven auth login and complete the browser flow. For CI environments, set the RIVEN_TOKEN environment variable instead.

Error: Docker daemon not running

The CLI requires Docker for building images. Start Docker Desktop or the Docker daemon:

Terminal
bash
# macOS / Windows
open -a Docker
 
# Linux
sudo systemctl start docker

Next Steps

You have successfully deployed your first Riven project. Here is what to explore next:

  • Configuration — Customize your project settings and environment variables.
  • Authentication — Set up tokens for CI and multi-org workflows.
  • AI Platform — Learn how to deploy and serve AI models.
  • CLI Reference — Explore all available CLI commands for proto management, deployments, and tasks.
PreviousInstallationNextConfiguration