Welcome to Tensorify Documentation

Tensorify is a visual workflow builder that generates executable Python code. Design AI workflows visually and export production-ready code that runs locally or in the cloud.

What is Tensorify?

Tensorify lets you build complex AI and data workflows using a visual drag-and-drop interface. Instead of writing boilerplate code, you:

1. Design visually - Drag plugins onto a canvas and connect them
2. Configure easily - Set parameters through intuitive UI forms
3. Export code - Get production-ready Python that you own
4. Run anywhere - Execute locally or on cloud GPUs

Key Features

Visual Workflow Builder

Build sophisticated AI pipelines without writing repetitive code. Our canvas interface lets you see the entire data flow at a glance.

40+ Pre-built Plugins

Access a growing library of plugins for:

• AI Models - OpenAI, Anthropic, Hugging Face
• Data Processing - CSV, JSON, PDF parsing
• Databases - PostgreSQL, MongoDB, Pinecone
• Integrations - Slack, SendGrid, webhooks

Code Generation

Every workflow compiles to clean, readable Python code. No vendor lock-in - you own and can modify the generated code.

Cloud Execution

Run workflows on Modal with automatic GPU provisioning. Scale from prototyping to production without infrastructure changes.

Quick Links

| Resource | Description | | -------------------------------- | -------------------------------------- | | Quick Start | Get up and running in 5 minutes | | Core Concepts | Understand workflows, nodes, and edges | | Plugin Catalog | Browse available plugins | | API Reference | REST API documentation |

Example Workflow

Here's a simple workflow that processes documents with AI:

# This code is auto-generated by Tensorify

from openai import OpenAI
import json

# Initialize OpenAI client
client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])

# Read input document
with open("input.txt", "r") as f:
    content = f.read()

# Process with GPT-4
response = client.chat.completions.create(
    model="gpt-4",
    messages=[
        {"role": "system", "content": "Summarize the following document."},
        {"role": "user", "content": content}
    ]
)

# Save result
summary = response.choices[0].message.content
print(summary)

Getting Help

• Discord Community - Join our community for real-time help
• GitHub Issues - Report bugs and request features
• Email Support - [email protected] for enterprise inquiries

---

Ready to get started? Head to the Quick Start Guide →