Overview
End-to-end builds with Docling for IBM watsonx
Tutorials take you from an empty project to a working application. Each one lists its prerequisites and how long it takes, and ends with a way to check that everything works. You can reproduce all of them with your own accounts.
Get the code
All the code lives in the docling-ibm-examples repository, with one folder per tutorial:
git clone https://github.com/IBM/docling-ibm-examples.gitRAG in a notebook
15 min · The smallest possible RAG application: convert with the service, chunk, embed and answer with watsonx.ai.
Ingest a document library into a vector database
15 min · One API call from document URLs into AstraDB, with no glue code.
Streaming pipelines with Confluent Cloud
20 min · Stream Docling chunks into Confluent Cloud, then clean, redact and deduplicate them with Flink SQL. No scripts.
Docling in IBM Bob
30 min · Give IBM Bob document understanding with an agent skill. Fill a compliance form from a PDF, a Word file and an e-book, without writing code.
Which one should I start with?
| If you want to… | Start with | ||
|---|---|---|---|
| Learn the RAG building blocks | RAG in a notebook | 15 min | |
| Load an existing document collection into search | Vector database ingestion | 15 min | |
| Process new documents continuously as they arrive | Streaming pipelines with Confluent Cloud | 20 min | |
| Work with documents from an AI coding assistant | Docling in IBM Bob | 30 min |
Command Line
Convert files, folders and URLs with Docling for IBM watsonx from the terminal using docling convert-remote
Streaming Document Pipelines with Confluent Cloud
Stream Docling for IBM watsonx chunks into Confluent Cloud and clean, redact and deduplicate them with Flink SQL, ready for search and RAG