TensorifyDocs

Installation

Tensorify uses a web app for editing and a local CLI for execution.

What you need installed

For a local-first workflow, you need:

  • a browser for app.tensorify.io
  • the Tensorify CLI on your machine
  • Python available for local runtime/export workflows

Web app

The editor itself is browser-based. You do not need to install the editor locally just to build workflows visually.

Local execution requirements

Tensorify depends on local execution commands such as:

tensorify watch <workflowId>
tensorify run <workflowId>
tensorify export <workflowId>

Python

Verify Python is available:

python --version

Use Python 3.10+ where possible.

Recommended local environment

Use a virtual environment for exported or cloned workflow dependencies:

python -m venv tensorify-env
source tensorify-env/bin/activate

Windows:

tensorify-env\Scripts\activate

Environment variables

Many workflows depend on local environment variables for external services.

Examples:

export STRIPE_WEBHOOK_SECRET="..."
export API_BASE_URL="https://internal.example.com"

Use whatever variables your workflow needs. The execution model is local-first, giving you full control over your data and environment.

Docker

Docker can be useful for reproducible local workflow environments.

Next