surface↓classification↓engagement_llm
Classificationdimension · returns varchar

ENGAGEMENT_LLM

LLM-backed engagement-level prediction (escape hatch for ENGAGEMENT)

Per-row classifier — stable across GROUP BY.

classificationllmllm-escape-hatchtext

Arguments

nametypedescription
textVARCHAR—
platformVARCHAR—

About

LLM-backed alternative to ENGAGEMENT for predicting likely social- media engagement level. Reach for this when the base NLI operator can't see platform-specific signals — subtle humor, insider references, emoji-heavy posts, hashtag games, or platform idioms that don't read as "engaging" in isolation. Especially helpful when pairing with a platform hint: a joke that works on Twitter may fall flat on LinkedIn, and LLM reasoning picks up that context more reliably than zero-shot NLI against a template. For routine engagement bucketing on standard post formats, ENGAGEMENT is far cheaper and fast enough for dashboard-scale workloads.

Examples

LLM escape hatch predicts engagement

SELECT
  engagement_llm ('Must read! Share with everyone!')

Nearby rabbit holes

same domain
dimension

AUDIENCE

Identify target audience via zero-shot NLI

classificationnlispecialist-zootext+1
dimension

AUDIENCE_LLM

LLM-backed audience identification (escape hatch for AUDIENCE)

classificationllmllm-escape-hatchtext
dimension

AUTHENTICITY

Assess authenticity of content via zero-shot NLI

classificationnlispecialist-zootext
dimension

AUTHENTICITY_LLM

LLM-backed authenticity assessment (escape hatch for AUTHENTICITY)

classificationllmllm-escape-hatchtext
dimension

BUCKET

Classify text into user-specified buckets via zero-shot NLI

classificationnlispecialist-zootext
dimension

BUCKET_LLM

LLM-backed bucketing (escape hatch for BUCKET)

classificationllmllm-escape-hatchtext
Climb back to The Looking Glass