Skip to content
Metatext Decide — typed decisions for your agents, with a probability on every answerGet early access
Metatext
Sign Up

Jevflake

dbt package and Terraform module adding jev_noul, jev_choice, jev_score and jev_ask SQL functions to Snowflake.

Librarybeta
View on GitHub ↗

Overview

Jevflake lets Snowflake ask Jev typed questions about rows so that the answer comes back as a column you can filter, join and test. Because Jev returns a number or label with a probability rather than text, it fits naturally into SQL. The dbt package sets up Snowflake's external access so it can call the Jev API, creates SQL functions (jev_noul, jev_choice, jev_score and a general jev_ask), and provides dbt macros and tests so answers are stored once, reused and checked like any other model. Teams that manage Snowflake with Terraform can use the included module for setup and function creation instead.

The README explains the three question types, including that a Score is the probability-weighted average of level numbers, so three levels give a score from 0 to 2. Prerequisites include a Snowflake account with external access enabled (off by default on trials), a role able to create an integration, a TypeSafe key, and dbt 1.8 or newer with dbt-snowflake, tested on 1.12. The Python function relies on pandas and requests provided by Snowflake. It is not affiliated with TypeSafe, Snowflake or dbt Labs.

What the decision model decides

TypeQuestionAnswers
Yes / noRow-level yes/no question via jev_noul—
ChoiceRow-level classification via jev_choice—
ScoreRow-level rating via jev_score—

How it works

  1. Configure Snowflake external access and integration
  2. Create Jev SQL functions via dbt or Terraform
  3. Use functions in dbt models; store answers once
  4. Test answers with dbt tests

Strengths

  • Answers become ordinary columns
  • Stored once and reused to control cost
  • dbt tests on model outputs

Caveats

  • Requires Snowflake external access
  • Needs admin-level role to set up

At a glance

Type
Library
Status
beta
Language
SQL
Runtime
Snowflake + dbt / Terraform
Has code
yes
Has tests
yes

Use cases and integrations

How curated lists describe it

  • A dbt package and Terraform module that let Snowflake ask a typed question about a row, so the answer comes back as a column you can filter, join and test.

    MrJev/awesome-jev

Frequently asked questions

What SQL functions does it add?

jev_noul, jev_choice, jev_score and jev_ask.

Can I avoid repeat costs?

Yes, dbt macros store answers once for reuse.

Does it work on Snowflake trials?

Only if external access is enabled, which is off by default on trials.

Related projects

  • pg-jev

    PostgreSQL extension with jev(), jev_prob, jev_choice and jev_score to filter, rank and classify rows by plain-language conditions.

    PluginbetaC
  • dbt\_jev

    dbt package whose macros run Jev Choice, Noul and Score over SQL values in DuckDB and ClickHouse models.

    ProjectbetaSQL (Jinja)
  • duckdb-jev

    DuckDB extension that asks Jev about every row in SQL and returns answers as ENUM, DOUBLE or STRUCT columns.

    PluginbetaC++
  • jevotron

    CLI that scans tabular, structured and text files or databases and asks Jev which fields look anomalous, building a review queue.

    CLIbetaPython
  • A DuckDB extension for row classification

    DuckDB extension that labels rows in CSV, Parquet or DuckDB tables with Jev, reported at about ten seconds per thousand rows.

    Discussionprototype
  • mysql-ailike

    Native MySQL plugin adding an AILIKE operator that asks Jev whether a value satisfies a natural-language condition.

    PluginbetaC++