Docs
OfficialJev Practices & Patterns
Advanced: Structure
Instructions, option descriptions, score levels, and noul criteria all accept JSON, so you can label question parts, pass schemas and taxonomies, and keep candidate paths alive.
- Format
- Docs
- Reading time
- 2 min
- Published by
- TypeSafe AI
- Last verified
- Sep 2026
Works forJev for Scoring
Used inJev for Software & DevTools
Highlights
- Every field is an entry type: string, object, array, or null.
- Reference state fields by key name in backticks, such as `potential_duplicate`.
- Structured options can state what an option covers, what it is not_for, and examples.
- Taxonomy walks pass a subtree per option, so every candidate path stays available in one call.
Quickstart
json
"instructions": {
"question": "Does the claimed sender identity conflict with the sending domain?",
"compare": ["ticket.sender.display_name", "ticket.sender.email"],
"focus": "Compare the named organization with the email domain."
}Watch out
Subtrees grow fast; trim a branch to its direct children plus a sample of leaves to keep the request affordable.
Questions about this entry
Short answers drawn from the linked project or page.
Yes. Every field is an entry type — string, object, array, or null — so instructions, options, score levels, and noul criteria all accept JSON. Taxonomy walks pass a subtree per option so every candidate path stays available in one call.
They grow fast and inflate the input bill. The page's advice is to trim a branch to its direct children plus a sample of leaves, and to reference state fields by key name in backticks such as `potential_duplicate`.
A string, object, array, or null. A single field object can describe the field being checked, with keys like question, compare, and focus, and each question refers to it by key.
It asks one Choice per level, passing each option's subtree as its value so the model sees what lives under a branch before committing; a beam search keeps several candidate paths alive when probabilities are close.
Option descriptions can be structured objects acting as a JSON rubric that states what each option does and does not cover, sharpening the boundary between options.
They are optional, but when the yes/no boundary is subtle, structured true and false descriptions pin it down with a definition and examples on each side.
Yes. The page says you can loop over potential records and build one structured question per field, all sent in a single call, as the SDE cascade cookbook does.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
X: How the agent stack moved in Jev's first three days
YouTube: Diogo Almeida tells swyx what System One is, why RLHF failed his automation goal, and why AI should eventually disappear.
X: The 10-step roadmap, summarised
X: Giving your agents a decision brain
More Jev resources like this
Architectural patterns for System One systems: speculative fan-out, confidence-gated routing, composite scoring, and intent routing.
Jev Practices & PatternsDocs#official#patterns#architecture
Official
Rate content against 2 to 10 ordered levels. Returns a probability-weighted score that can land between levels, per-level probabilities, and confidence, with the arithmetic left to code.
Jev Practices & PatternsDocs#official#primitives#score
Official
Break a judgment into atomic Score questions, normalize each by its top level, and combine them with weights you control in code, as in the resume example's four scored dimensions.
Jev Practices & PatternsDocs#official#patterns#score
Official
Ask everything the system might need in one request, then let code throw away what it does not use. The ticket-triage example sends a category plus four speculative questions in parallel.
Jev Practices & PatternsDocs#official#patterns#batching
Official
Classify intent with a Choice and complexity with a Score, then send each branch to deterministic code, a specialist model, or a person, with a 0.5 intent-confidence floor.
Jev Practices & PatternsDocs#official#patterns#routing
Official
The three TypeSafe question types, the typed answers they return, how to choose between them, and how to ask several in a single call.
Jev Practices & PatternsDocs#official#primitives#choice
Official
From the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.