Case Study: RTL Nederlands achieves scalable, data-driven video discoverability with Pachyderm

RTL Nederlands Relies on Pachyderm’s Scalable, Data-Driven Machine Learning Pipeline to Make Broadcast Video Content More Discoverable

RTL Nederlands, part of Europe’s largest broadcast group, needed to apply AI to vast amounts of unstructured video to make content more valuable and discoverable for millions of viewers. To automate labor‑intensive tasks like thumbnail and trailer selection, ad insertion, and multimodal tagging, the team selected Pachyderm for scalable, reproducible, data-driven ML pipelines.

Using Pachyderm’s pipeline orchestration and data versioning, RTL Nederlands broke complex video tasks into reusable subtasks (frame extraction, shot segmentation, speech detection, NLP tagging) and automated end-to-end workflows triggered by new data. The approach enabled parallel, incremental processing that lowered compute and storage costs, sped debugging via immutable lineage, and saved thousands of work hours per year—demonstrating Pachyderm’s ability to scale and reliably operationalize video AI.

RTL Nederlands

Vincent Koops

Senior Data Scientist