Add docs for Tetra launch: overview and quickstart

[muhsinking]

Add Tetra Documentation

This PR adds documentation for Tetra, a Python SDK that streamlines AI workflow development and deployment on RunPod's Serverless infrastructure.

Added Files

1. docs/tetra/overview.md

   • Introduces Tetra and its key benefits
   • Explains core concepts including resource configuration, remote functions, and data transfer
   • Details common use cases and how Tetra works under the hood

2. docs/tetra/get-started.md

   • Step-by-step tutorial for setting up Tetra
   • Covers environment setup, API key configuration, and basic usage
   • Includes a practical example of GPU-based matrix operations
   • Demonstrates both single and parallel operation execution
   • TODO BEFORE MERGING: Currently the config code example usestemplateId="[YOUR_TEMPLATE_ID]", but lacks steps for where to get this ID.

Key Features Documented

• Simplified workflow development using Python decorators
• Resource configuration through ServerlessResource objects
• Remote function execution with dependency management
• Asynchronous execution and parallel processing
• Data transfer between local and remote environments
• GPU and CPU resource allocation

Example Code

docs/tetra/get-started.md includes a practical example demonstrating:

• Basic matrix operations on GPU using NumPy
• Parallel processing using asyncio.gather()
• Resource configuration for a GPU endpoint
• Dependency management for remote functions

Code sample is mirrored here: runpod/tetra-rp#6

Testing

All code examples have been tested and verified to work on macOS with Python 3.12.

[vercel]

The latest updates on your projects. Learn more about Vercel for Git ↗︎

Name	Status	Preview	Comments	Updated (UTC)
docs	✅ Ready (Inspect)	Visit Preview	💬 Add feedback	Apr 24, 2025 9:18pm