UnDoc Document AI


UnDoc - Automate your document processing

UnDoc is an artificial intelligence-powered solution for automatic document processing. Allowing you to extract valuable business insights from complex data through real-time document processing. We use machine learning to classify documents, extract information from them, and embed this into your back-end systems. Resulting in improved operational efficiency, with a typical Return on Investment within one year. 

Request a demo
Eva Rapid Application Devevlopment

Rapid Application Development

Eva is a Productivity Improvement Framework for Application Development on Google Cloud It gives you the capability to accelerate the building of advanced Enterprise Business Applications at scale on Google Cloud through smart requirements gathering, instant prototyping, and rapid code generation.

Eva Rapid Application Devevlopment

Rapid Application Development

Eva is a Productivity Improvement Framework for Application Development on Google Cloud It gives you the capability to accelerate the building of advanced Enterprise Business Applications at scale on Google Cloud through smart requirements gathering, instant prototyping, and rapid code generation.

Unlocking value from documents

Your challenge

Many of your documents are received in email, pdf or on paper. Processing the information from these documents into your ERP or TMS is often done manually.  This is cumbersome, time-consuming and costly. 

Documents include for example:

UnDoc document AI your challenge
UnDoc document AI scanning

Capturing & digitizing at scale​

UnDoc will automatically open mailboxes for incoming documents, or retrieve digitized documents from other locations  to process them for document conversion based on Google Cloud’s OCR technology. 

Automated & accurate data extraction

Once received, UnDoc goes to work to make sense of unstructured data in these documents. Categorizing documents per type if needed, and extracting data as per the pre-defined relevant document fields (e.g. invoice line amount). That means you no longer have to map data to these fields manually.

UnDoc is optimized for advanced extraction at both header & line level which allows for extracting complete tables from documents, including for example:

Embedded into back-end & workflows

UnDoc provides a verification app to review and give corrections (human-in-the-loop) and enables self-learning for the machine models. The user is facilitated with dashboards for monitoring and performance reporting. The validated extracted data is fed into back-end systems like ERP or TMS realtime via an API.

Jumpstarting your automated document processing solution

Based on your unique need and use-case Vanenburg will develop an end-to-end document processing solution using customizable UnDoc components. This solution allows you to extract structured content from complex unstructured documents using Google’s industry-leading Artifical Intelligence & Machine Learning technologies and Vanenburg’s “building blocks”. 

 Components of the UnDoc solution include: 

Document capture

Scans mailbox / cloud storage for incoming documents and sends them for conversion 

Receives response and stores/routes them to required destinations 

Conversion service

OCR / Google DocAI

Custom AI/ML models for content classification and entity recognition 

Knowledge stores for data enrichment and validation 

Verification app

Verification desk to review and provide corrections 

Feedback loop to ML models for continuous learning

Views / Search / Reports

Dashboards

Prediction and extraction logging

Conversion service performance monitoring

Reporting & alerting

Business benefits

The UnDoc solution is optimally geared towards use cases requiring a specific and non-standard approach (e.g. complex documents with a high number of fields to be extracted at changing locations across the documents with high accuracy).  These less straightforward use cases are not handled well by default OCR solutions. 

Our customers further include companies that are looking to acquire a competitive advantage in this space (build own IP).  Or where high volumes of document to be processed warrant the investment in a dedicated solution.

Accurate

>90% automation rate.  With self-learning via verification app

Easy to use

End user friendly with exception workflows, and dashboards for insights

ROI < 1 year 

Save on cost of manual work or an inaccurate OCR solution with high license cost 

Fast implementation

Start quickly. Implement a solution optimized for you in ~6 weeks

Integrated

Truly automated solution, with data processed into your back-end systems 

Why Vanenburg?

Work with a team of experts that is global trusted partner of Google Cloud DocAI

Dedicated AI/ML practice next to App Development, Data & CloudOps capabilities

Leverage our ready to use components for a custom solution with attractive ROI

Experience with implementing this at enterprise scale

Rapid app development tooling to build apps on Google Cloud

Experience to integrate the solution into back-end systems

Trusted by our customers

Share your documents for a complementary demo! 

Automate your processes today! Share 5 example documents you would like to digitize and mark the fields that are important to you. You will receive a complementary demo on how we digitize these documents rapidly. 

Request demo