IBM Developer: Data Prep Kit for LLM Data

Understand the project

Practical overview

Use the DPK learning material when planning or learning a data-preparation workflow for an LLM application, including RAG or fine-tuning.

Developers preparing unstructured data for LLM applications.
Practitioners learning data preparation for RAG, fine-tuning or agentic workflows.

Extract material from sources such as PDFs, HTML and code.
Convert extracted data into tokens suitable for LLMs and vector databases.
Learn how data preparation fits into RAG and fine-tuning pipelines.
Explore pre-built transforms and the creation of custom data-preparation modules.

Before you choose

The learning path assumes basic Python knowledge.
Google Colab is the stated notebook environment for the learning path.
The material covers data extraction, token preparation, transforms and application workflows, but does not specify DPK’s implementation, interfaces or runtime.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Learn the documented data-preparation concepts and workflows for LLM applications.

Open the Data Prep Kit learning path.
Work through its introduction to DPK and data ingestion.
Study the sections on preparing data for fine-tuning, RAG and agentic workflows.
Continue to the modules on pre-built transforms, custom modules and scaling workflows.

A guided introduction to preparing unstructured data for LLM application workflows.

The learning path describes topics and learning objectives; these steps do not establish a working DPK environment or execute a data-preparation pipeline.

Evidence and transparency

Source record details

Source availability: Active
Sources checked: 12 October 2026
Original website title: Data Prep Kit (DPK) - IBM Developer · Official homepage title · retained in its source language