Atelier

Atelier is the referee engine on the Signals lattice (gRPC :50251). It is an agentic classification workbench: a gRPC core and a React canvas where analysts steer classification across thousands of columns.

The pipeline combines six evidence sources — embedding similarity, gradient-boosted prediction, regex pattern detection, column-name matching, short-text SVM, and an LLM convergence agent — through Dempster–Shafer belief fusion. Each source contributes a mass function over a restricted frame of discernment; combination yields a belief interval at every node of the hierarchy. Interval width is epistemic uncertainty made legible.

A convergence agent targets the columns where the belief gap is widest, gathers additional context, and proposes a reclassification. Embeddings are explored on a 2-D atlas — a derivative of Apple’s embedding-atlas modified for this workflow.

Atelier classifies into the shared SKOS vocabulary published by SDG corpora. It is the short-iteration counterpart to Ægir: per-column, agent-assisted convergence, with outcomes stored for Signals governance.