UnderDoc is a simple, easy to use and cost effective document understanding platform.

With the latest GenAI technologies, many LLMs can now accept multi-modal inputs (e.g. images) and produce structured output base user instructions and the content/layout appeared on images.

Our mission is to leverage these exciting technologies to re-imagine how document understanding can be done.

• Our first service is an API SaaS that can use to extract structured data from any expense images (e.g. invoice, receipt, demand notes, etc.)
• We also provide professional services for customers who want to extract structured data from their documents and images
• Our focus is in Asia market and we test our platform with a combination of English and Asian (e.g. Chinese, Japanese, etc.) languages

To get start, simply visit our UnderDoc Developer Portal (https://dev-portal.underdoc.io), signup for an account (free and no credit card required).

In our Developer Portal, there is a Playground that you can try out your own images and view the extracted output.

For our Expense Document Understanding API, you can see the API Spec at https://api.underdoc.io/redoc

To further simplify developer experience in using UnderDoc, we also offered a Python SDK.

To use the SDK, first signup and get a key at our Developer Portal.

Then in your Python environment, install our SDK. The following is an example using pip:

pip install underdoc

You can refer to the document and examples in our publib Github repo: https://github.com/under-doc/underdoc-python-sdk

Visit our website (https://www.underdoc.io) for more information, sample images and their extracted output, as well as our pricing plan.

For any questions, feel free to contact us at enquiry@underdoc.io