Introducing workflow, the NLP orchestration engine

We're thrilled to introduce workflow, our answer to questions such as:

• I don't have collected data yet, what can I do?
• I want to continuously feed data into refinery, but how?
• I want to automate a process once I've built the API, how can I best do so?

In our platform, workflow is the beginning and the end of the user experience.

You can start using workflow to collect data from different data sources, e.g. Google mail or spreadsheets, or from a large-language-model that generates data for you given some example record and instructions.

The data in workflow is moving; you can define steps in workflows, e.g. to flow data from a source store to a target store. Inbetween that steps, you can use gates to use your refinery projects easily for enrichment of textual data.

With workflow, you'll be able to implement a wide set of use cases, such as building training data. We're sure you'll love it!