Pipelinepipeline · returns table

ENRICH

Add LLM-computed columns to query results

Table-in, table-out — composes downstream of SELECTs.

pipelinellmpipeline-composabletext

Syntax

THEN ENRICH {{ prompt }}

THEN ENRICH({{ prompt }})

Arguments

name	type	description
prompt	VARCHAR	—
_table	TABLE	—

About

PIPELINE cascade for enriching query results with LLM-computed columns. Used with THEN ENRICH syntax for adding semantic columns: SELECT * FROM products THEN ENRICH('add sentiment score and category') Uses LLM to compute additional columns for each row based on the existing data and the enrichment specification.

Examples

Preserves original rows while adding the requested enrichment column

SELECT
  *
FROM
  (
    VALUES
      ('Great product'),
      ('Terrible experience')
  ) AS t (review) THEN ENRICH (
    'add a sentiment_label column with values positive or negative'
  ) THEN PYTHON (
    'result = pd.DataFrame({"ok":["sentiment_label" in df.columns and len(df) == 2]})'
  )

Nearby rabbit holes

same domain

pipeline

ADD_STYLES

Apply theme styling to a chart specification

pipelinellmpipeline-composabletext

pipeline

ANALYZE

Analyze query results with LLM based on a prompt

pipelinellmpipeline-composabletext

pipeline

DEDUPE

Remove duplicate rows

pipelinellmpipeline-composabletext

pipeline

FILTER

Filter query results using LLM-based semantic matching

pipelinellmpipeline-composabletext

pipeline

GROUP

Group by column and aggregate another

pipelinellmpipeline-composabletext

pipeline

INVESTIGATE

Investigative analysis - explores related data to answer questions

pipelinellmpipeline-composabletext

Climb back to The Looking Glass

surface↓pipeline↓enrich

Pipelinepipeline · returns table

ENRICH

Add LLM-computed columns to query results

Table-in, table-out — composes downstream of SELECTs.

pipelinellmpipeline-composabletext

Syntax

THEN ENRICH {{ prompt }}

THEN ENRICH({{ prompt }})

Arguments

name	type	description
prompt	VARCHAR	—
_table	TABLE	—

About

Examples

Preserves original rows while adding the requested enrichment column

SELECT
  *
FROM
  (
    VALUES
      ('Great product'),
      ('Terrible experience')
  ) AS t (review) THEN ENRICH (
    'add a sentiment_label column with values positive or negative'
  ) THEN PYTHON (
    'result = pd.DataFrame({"ok":["sentiment_label" in df.columns and len(df) == 2]})'
  )