Use casesClassification pipelines

Content tagging and catalogue normalisation

Marketplaces normalising millions of messy supplier feeds into one schema.

The problem

Each supplier describes the same product differently: "TV", "Television", "55in LED". Marketplaces need a canonical category tree, attribute set, and compliance flags before items go live. Rules engines break on edge cases; pure LLM tagging drifts week to week.

Why Ecko

Pipeline stages
Use |> to chain normalise, validate, enrich, and publish. The AI step is one stage in a typed pipeline, not a separate microservice.
Batch economics
Per-record cost() makes margin visible when a supplier sends fifty thousand SKUs nightly.
CSV and SQL IO
Read supplier feeds and write catalogue tables with stdlib modules - no Spark cluster for a job that fits on one machine.

In practice

catalog.ecko
fn normalise(row: SupplierRow) -> CanonicalItem {
  row
    |> clean_title
    |> ai_tag
    |> validate_schema
}

Further reading

Related cases

All 32 use cases →

Try it on your workload.

Install Ecko Open the playground