foundations 8 min Updated Sep 19, 2026

System One vs an LLM

Jev selects typed decisions with probabilities. LLMs generate text. Use both — do not force one job onto the other.

Two jobs

LLMJev (System One)
OutputStrings / prose / codeChoice, Score, Boolean (+ probabilities)
SamplingToken by tokenParallel answers over declared questions
Failure modeHallucinated or off-schema textCannot leave the schema you defined (product claim)
Sweet spotWriting, planning, tool-argument proseRoute, score, verify, guardrail

TypeSafe’s launch post frames Jev as a frontier-intelligence function call: unstructured state in, typed probabilistic decisions out.

Composition pattern

  1. LLM drafts or gathers context.
  2. Jev decides (queue, score, allow/deny).
  3. Code branches on confidence.
  4. LLM writes only when a human-readable artifact is required.