Lists & resources

awesome-jev-usecases

@aliaihub★ 7NOASSERTIONupdated 2026-09-18▶ runnable

Evidence-backed use cases, patterns, and guidance for building with Jev, TypeSafe AI's System One model. Every claim is labeled and sourced.

aliaihub/awesome-jev-usecases ↗

Where it calls Jev

from typesafe_sdk import TypeSafeClient  # imported lazily

examples/python/_client.py:324

The link points at the commit we read, so the line number still holds.

What it asks Jev

These question sets are lifted from this project's source exactly as written.

from examples/python/01_routing_triage.py:38

  1. topicchoice

    question

    • what — Does not fit any of the above
    • not_for — Charges or order tracking

Run it on your own text

The code
import { experimental_evaluate as evaluate } from 'ai';

const { answers } = await evaluate({
  model: 'typesafe-ai/jev',
  state,
  questions: {
    topic: { type: 'choice', instructions: 'question',
      criteria: { what: 'Does not fit any of the above', not_for: 'Charges or order tracking' } },
  },
});

from examples/python/02_composite_and_fanout.py:66

  1. python_depthscore

    Depth of Python experience shown

    • 0 — None mentioned
    • 1 — Mentioned, no detail
    • 2 — Used in projects
    • 3 — Primary language
    • 4 — Deep expertise: architecture, performance
  2. team_leadershipscore

    Experience leading engineering teams

    • 0 — None
    • 1 — Informal mentorship
    • 2 — Led a small team
    • 3 — Managed direct reports
    • 4 — Managed multiple teams
  3. system_designscore

    Experience designing distributed systems

    • 0 — None mentioned
    • 1 — Contributed to discussions
    • 2 — Designed components
    • 3 — Owned a system's architecture
    • 4 — Designed at scale across domains

Run it on your own text

The code
import { experimental_evaluate as evaluate } from 'ai';

const { answers } = await evaluate({
  model: 'typesafe-ai/jev',
  state,
  questions: {
    python_depth: { type: 'score', instructions: 'Depth of Python experience shown',
      criteria: ['None mentioned', 'Mentioned, no detail', 'Used in projects', 'Primary language', 'Deep expertise: architecture, performance'] },
    team_leadership: { type: 'score', instructions: 'Experience leading engineering teams',
      criteria: ['None', 'Informal mentorship', 'Led a small team', 'Managed direct reports', 'Managed multiple teams'] },
    system_design: { type: 'score', instructions: 'Experience designing distributed systems',
      criteria: ['None mentioned', 'Contributed to discussions', 'Designed components', 'Owned a system\'s architecture', 'Designed at scale across domains'] },
  },
});

from examples/python/04_cascade.py:38

  1. intentchoice

    Primary intent of this message

    • order_status — Asking about an existing order
    • product_question — Asking about a product
    • return_exchange — Wants to return or exchange
    • complaint — Unhappy, wants resolution
    • other — None of the above
  2. complexityscore

    How complex is this to resolve?

    • 0 — Simple lookup or standard procedure
    • 1 — Requires judgment or multiple steps
    • 2 — Unusual edge case, escalation needed

Run it on your own text

The code
import { experimental_evaluate as evaluate } from 'ai';

const { answers } = await evaluate({
  model: 'typesafe-ai/jev',
  state,
  questions: {
    intent: { type: 'choice', instructions: 'Primary intent of this message',
      criteria: { order_status: 'Asking about an existing order', product_question: 'Asking about a product', return_exchange: 'Wants to return or exchange', complaint: 'Unhappy, wants resolution', other: 'None of the above' } },
    complexity: { type: 'score', instructions: 'How complex is this to resolve?',
      criteria: ['Simple lookup or standard procedure', 'Requires judgment or multiple steps', 'Unusual edge case, escalation needed'] },
  },
});

from examples/python/05_retrieve_then_judge.py:53

  1. evidence_strengthscore

    How strong is the causal evidence presented?

    • 0 — Anecdotal or preclinical
    • 1 — Observational
    • 2 — Single randomised trial
    • 3 — Meta-analysis of randomised trials

Run it on your own text

The code
import { experimental_evaluate as evaluate } from 'ai';

const { answers } = await evaluate({
  model: 'typesafe-ai/jev',
  state,
  questions: {
    evidence_strength: { type: 'score', instructions: 'How strong is the causal evidence presented?',
      criteria: ['Anecdotal or preclinical', 'Observational', 'Single randomised trial', 'Meta-analysis of randomised trials'] },
  },
});

Other projects in this category