Everyrow icon

Everyrow

AI-powered data operations: rank, dedupe, merge, screen, and agent workflows

Actions7

Data Operation → Rank

AI-generated

Summary

Rank rows by scoring them according to user-defined criteria, extracting and sorting by a specified field in ascending or descending order.

Inputs

  • Session Name — Name for the Everyrow session used for tracking in the Everyrow dashboard.
  • Task Description (required) — Description of the criteria to use for ranking the rows.
  • Field Name (required) — Name of the field in which the ranking score will be extracted and sorted.
  • Sort Order — Boolean to determine if sorting is in ascending order (lowest first). Defaults to true.
  • Response Schema (JSON) — Optional JSON Schema defining the response format. If omitted, a default schema containing the specified sort field is used.

Output shape

a single record representing the submitted task with properties including task_id, session_id, status, and operation name ('rank').

The response includes task metadata immediately after submission. It does not return ranked data directly but provides identifiers to later retrieve results via separate task status and result operations.

Examples

Example 1: Rank companies by relevance to AI infrastructure

Set Task Description to 'Score each company by their relevance to AI infrastructure', Field Name to 'score', Sort Order to ascending (true). Optionally provide a Response Schema to define the fields returned including 'score' and 'reasoning'.

Links

Discussion