Classify (Classification/NLU) — foundations RFC

Status

*ccepted*— 20260509. Sector bootstrap (skeleton + 5 impl tickets) landed as part of /k-go services/ai audit wave (Modo C). Q1 resolved: fastText for language, BGEclass for sentiment + multilabel, LLM via gateway for zeroshot tail. Q2 resolved: pertenant taxonomies stored in this sector under kdb-doc, versioned; fewshot routing when ≥5 examples per intent, zeroshot otherwise.

Summary

Classificação — intent, sentiment, topic, language detection. guard/ faz só safety; este é genérico.

Motivation

Classificação é primitiva ML clássica que LLM faz mal e caro. Modelos pequenos resolvem 100x mais barato. Sem service, cada produto chama gateway com LLM grande.

Scope

In

• Zero-shot classify (LLM fallback)
• Fewshot finetuned
• Language detect
• Sentiment
• Intent

Out (yet)

• Custom training (escopo training/)

Initial design

Surfaces

• backend/ — Go API + worker (modelos pequenos local)
• app/ — não aplicável v1

Key APIs

• POST /v1/classify/labels — multi-label
• POST /v1/classify/sentiment — posnegneutral
• POST /v1/classify/language — detect
• POST /v1/classify/intent — intent recognition

Dependencies

• services/ai/runtime — modelos pequenos local
• services/ai/gateway — fallback LLM
• services/ai/embed — semantic similarity

Relation to existing sectors

• Distinto de guard/ (safety classifier)
• Consumido por chat-adapter (intent), ingest pipelines

Selfhostedfirst analysis (5 gates)

• *1 Feature parity* zero
• *2 Performance* N/A
• *3 Stability* N/A
• *4 Capability* BERT-tiny, fastText viáveis
• *5 Critical-path readiness* destrava ingest pipelines + chat routing

Open questions

• Q1: Default models — fastText (light) ou BGE-class (heavier)?
• Q2: Custom labels — onde armazena?

Next steps

1. Ratificar esta RFC (1 round de comments).
2. Criar sector dir services/ai/classify/ com koder.toml, README.md, skeleton.
3. Abrir tickets de implementação em services/ai/classify/backlog/pending/.
4. Registrar em meta/docs/stack/registries/self-hosted-pairs.md se substituir externo.