Story
Dataiku
dunnhumbydunnhumby Uses MLOps for Event Streams
dunnhumby, an information technology organization in the United States, uses MLOps from Dataiku to support real-time operations for platform engineers.
MLOps became the place dunnhumby uses to move operational events between systems as they happen.
“Information Technology work at dunnhumby spans more than one site, even when headquarters sits in the United States.”
“Platform engineers asked for a shared way to run real-time operations without freezing local judgment.”
Independent write-up. Figures come from public sources or are illustrative.
Value results
| Category | Value result |
|---|---|
| Capability | Named workflow replaces ad hoc routing for event streams |
| Capability | Real-time operations stays visible to adjacent teams through MLOps |
| Capability | Platform engineers work from the same MLOps record for event streams |
Story
Information Technology work at dunnhumby spans more than one site, even when headquarters sits in the United States. Event streams was splitting across regional habits. Platform engineers asked for a shared way to run real-time operations without freezing local judgment.
Dataiku (MLOps) is what they standardized on. Dataiku is a universal AI platform that lets data teams and analysts collaborate on pipelines, models, and governed AI applications. dunnhumby uses it as the system of record for event streams, with platform engineers as the primary operators and other groups coming in through the same queue.
Leaders get a picture they can actually walk. Teams get fewer mystery statuses. The story is about operating change, not an unpublished percentage.
