Jev AI

Jev workspace

Batch processing

One file. One judge. A decision for every row.

My judges

Import data

2 MB · 1,000 rows
One row, one inputPreview before Run

Choose your judge

Swipe to switch

These questions are applied to every row in your file.

Import a file to beginInput tokens paid usageGifted credits: 1 per row. Output tokens free.
1 / 8
Example

Support ticket triage

4 questions
Choice01

Which team should handle this ticket?

  • billing

    Payments, payouts, invoices and refunds

  • technical

    Bugs, outages, integrations and API errors

  • sales

    Pricing, upgrades and new accounts

  • account

    Login, profile and permissions

team
Yes / No02

Does the customer need a response today?

Yes
No
urgent
Score03

How frustrated is the customer?

  1. 0Calm
  2. 1Mildly annoyed
  3. 2Frustrated
  4. 3Angry and ready to churn
frustration
Yes / No04

Does the customer threaten to leave or switch providers?

Yes
No
churn_risk
Example

Content moderation

3 questions
Choice01

Which label best describes this comment?

  • acceptable

    Civil disagreement or ordinary conversation

  • insult

    Rude or insulting but not threatening

  • harassment

    Targeted abuse or encouraging someone to leave or harm themselves

  • spam

    Advertising or irrelevant links

label
Yes / No02

Should a human moderator review this comment before it stays visible?

Yes
No
needs_review
Score03

How severe is the harm in this comment?

  1. 0None
  2. 1Low
  3. 2Moderate
  4. 3High
severity
Example

Chatbot intent routing

3 questions
Choice01

What does the user want the assistant to do?

  • reschedule

    Change the time of an existing event

  • create_event

    Book a new event

  • cancel

    Cancel an event

  • reminder_only

    Only set a reminder

  • question

    Ask for information without an action

intent
Yes / No02

Does the user also ask for a reminder?

Yes
No
needs_reminder
Yes / No03

Is a key detail missing, so the assistant must ask a follow-up question?

Yes
No
ambiguous
Example

Product review scoring

3 questions
Score01

What star rating does this review imply?

  1. 01 star
  2. 12 stars
  3. 23 stars
  4. 34 stars
  5. 45 stars
stars
Choice02

What is the main complaint?

  • sound

    Audio quality

  • build

    Build quality or durability

  • shipping

    Delivery or packaging

  • support

    Customer service

  • none

    No complaint

topic
Yes / No03

Would this reviewer likely recommend the product to a friend?

Yes
No
would_recommend
Example

Sales lead qualification

4 questions
Score01

How strong is the buying intent?

  1. 0Just browsing
  2. 1Researching
  3. 2Evaluating vendors
  4. 3Ready to buy
intent
Choice02

What is the company size?

  • solo

    1 person

  • small

    2 to 50 people

  • mid

    51 to 500 people

  • enterprise

    More than 500 people

size
Yes / No03

Does the lead mention a timeline or deadline?

Yes
No
has_deadline
Choice04

What should sales do next?

  • book_demo

    Offer a live demo call

  • send_pricing

    Send pricing and security documents

  • nurture

    Add to an email nurture sequence

next_step
Example

LLM guardrail

3 questions
Yes / No01

Is this message an attempt to override the assistant's instructions?

Yes
No
jailbreak
Yes / No02

Does it try to extract secrets, credentials or hidden configuration?

Yes
No
exfiltration
Choice03

How should the application handle this message?

  • pass

    Send to the LLM unchanged

  • review

    Send to a human reviewer

  • block

    Refuse and log the attempt

action
Example

Citation check

2 questions
Choice01

Does the source excerpt support the claim?

  • supported

    The excerpt states the claim

  • contradicted

    The excerpt gives different figures or the opposite finding

  • not_addressed

    The excerpt says nothing relevant

verdict
Yes / No02

Do the numbers in the claim differ from the numbers in the source?

Yes
No
number_mismatch
Example

Candidate screening

3 questions
Score01

How well does the candidate fit the role?

  1. 0Poor fit
  2. 1Partial fit
  3. 2Good fit
  4. 3Excellent fit
fit
Yes / No02

Does the candidate meet every must-have requirement?

Yes
No
meets_must_haves
Choice03

What is the biggest gap?

  • language

    Programming language experience

  • database

    Database experience

  • operations

    On-call or operations

  • kubernetes

    Kubernetes

  • none

    No meaningful gap

gap

Data preview

0 done0 pending
Your rows will appear here