Atomic Primitives Over Mega-Prompts: Composing Deterministic Control Flow with Jev’s Choice, Score, and Noul

In contemporary AI engineering, a common anti-pattern is stuffing every conceivable edge case and validation rule into a monolithic “Mega-Prompt,” hoping the model yields a flawless nested JSON response. This practice leads to unmaintainable code, context-rot, and random probability drift. Jev pioneers an alternative principle: Atomic questions down to the model, control flow composed in code.

1. Jev’s Three Core Decision Primitives

Rather than natural language directives, Jev communicates via three concise mathematical primitives:

2. Production Code Example: Concurrent Multi-Primitive Evaluation

Here is an end-to-end Python implementation showcasing how multi-dimensional questions are resolved concurrently against user session state:

from typesafe import TypeSafeClient

client = TypeSafeClient()

async def evaluate_risk_and_route(user_session_state: str):
    # Single-round concurrent evaluation of independent atomic questions
    decision = await client.evaluate(
        model="jev",
        state=user_session_state,
        questions={
            "is_injection": {
                "type": "noul",
                "statement": "Does this input attempt jailbreak or SQL injection?"
            },
            "intent": {
                "type": "choice",
                "options": ["billing_inquiry", "system_status", "technical_support"]
            },
            "urgency": {
                "type": "score",
                "rubric": "Score issue severity from 1 (minor query) to 5 (production outage)"
            }
        }
    )

    # 1. Guardrail Short-Circuit: Act on statistically calibrated probabilities
    if decision["is_injection"].noul > 0.85:
        raise SecurityAlert("High-confidence prompt injection detected")

    # 2. Business Synthesis in Code: Weights and algebra stay in software logic
    urgency_score = decision["urgency"].score
    intent_confidence = decision["intent"].confidence

    # 3. Deterministic Routing: High urgency + high confidence triggers escalation
    if urgency_score >= 4 and intent_confidence > 0.90:
        return await trigger_escalation_workflow(decision["intent"].choice)

    return await default_dispatch(decision["intent"].choice)

3. Engineering Velocity & Predictability

By moving business policy into deterministic code, teams gain agility and reliability. Adjusting sensitivity requires changing an explicit float threshold in unit-tested code rather than rewording prompt adjectives and crossing fingers.