Welcome to the Tensorlake Cookbook! This repository contains a collection of sample notebooks, sample applications to help you get started with document AI and Tensrolake serverless apps.
-
This cookbook provides a range of small, focused examples demonstrating how to parse real-world documents with Tensorlake Document AI and use the Tensorlake serverless platform to deploy apps.
-
Each sample lives in its own directory and comes with its own README to help you understand, run, and extend it quickly.
To get started with any of the samples, follow these steps:
-
Clone the repository
git clone https://github.com/tensorlakeai/cookbooks cd cookbooks -
Navigate to a sample directory
For example, to open the Outlines + Tensorlake example:
cd using-outline-with-tensorlake -
Follow the sample’s README
Each sample directory has its own
README.mdwith detailed instructions for installing dependencies, setting environment variables, and running the notebook or app.
Here are the samples currently included in this repository:
| Sample | Description | Link |
|---|---|---|
| Using Outlines with Tensorlake (Schema-Enforced Invoices) | Parse an invoice PDF with Tensorlake, then use Outlines + OpenAI to generate JSON that must match a Pydantic Invoice schema (no malformed outputs). |
using-outline-with-tensorlake |
Contributions are welcome!
Typical ways to contribute:
- Add a new sample under a new directory (e.g.,
contracts-structured-extraction/) - Improve documentation or comments in existing notebooks
- Share bug fixes or small enhancements (via PR)
If you add a new sample, please:
-
Create a folder at the repo root, e.g.
my-new-sample/ -
Add:
- A notebook or application
- A
README.mdexplaining what it does and how to run it
-
Update the Sample Cookbooks table in this root
README.mdwith:- Sample name
- Short description
- Link to the folder
-
Raise a PR with changes.
This project is intended to be licensed under the MIT License
See the LICENSE file for full details once you’ve added it.