I want to implement an interactive workflow and therefore created a workshop application. The user can upload PDF files to a media set. This media set triggers a pipeline and finally results in the backing dataset of an Ontology Object Type, which is displayed in the workshop app. All of this works pretty well, but I would like to give the user some feedback once they uploaded the PDF until the processed new Object becomes visible in the workshop app (could take a couple of minutes for the pipelines to process the new PDF). Also, I would like to display the current state (e.g. PDF text extraction, information processing, aggregation, syncing to Ontology, …).
In regards to your second question about progress, you might find this link helpful: https://build.palantir.com/platform/9a99bbfb-cb1d-4347-a61e-8b4c07625023
For the first question, could you elaborate on the type of feedback you’d like to provide? Depending on your use case, this could range from simple status updates (e.g., ‘Processing PDF’ or ‘Syncing to Ontology’) to more detailed integrations, such as using an LLM to extract and display relevant data or logs dynamically.
I was looking for more general status updates e.g.
I upload a PDF via the workshop action to a media set (status: “PDF uploaded to media set”)
The first pipeline which does the PDF processing is triggered (status: “PDF processing running”)
The first pipeline is finished and the second pipeline is triggered (status: “Entity extraction running”)
The second pipeline is done and the resulting dataset is the backing dataset of an Ontology object type (status: “Ontology syncing”)
Once this is done I see my object in the workshop app with all the information from the PDF, meta data etc
I am curious about the set up to get the status information, to make the workflow more transparent to the user (as opposed to user uploads PDF and has to wait 5 minutes) and to improve the workshop app.