Use cases· Last updated 2026-09-21

RAG evidence collection with Jev

Retrievers hope. After retrieval, Jev marks relevance, contradiction, or injection; code keeps, flags, or drops passages before a generator sees them.

This unofficial page is the evidence collection slice of the RAG passage decisions pack. Intent: apply the Jev (TypeSafe System One) decision model to RAG passage decisions evidence collection. Primary search language: RAG Jev evidence collection. Confirm patterns on docs.typesafe.ai. This site does not sell, issue, or proxy TypeSafe keys. Use a credential you already have from the console or a documented gateway.

Independent angle (cover ≠ clone): Filter/rerank recipes with failure modes; cite vs generate boundary. We cover the intent, not a rival rerank-passages-score URL tree.

RAG use-case context

Evidence collection for RAG passage decisions happens before POST /v1/systemone. Jev does not browse your warehouse, retriever, or ESP. You gather the query + retrieved passage facts, filter them, then ask snap questions. This slice is where fan-out cost math belongs: batch questions, do not re-send state.

Hub: Classifying RAG passages. Compare, when the other tool is the real job: RAG pipelines.

Evidence Collection inputs

Collect:

• The user query (verbatim)
• One passage (or a short pair) the questions name
• Optional trust tag if policy differs by corpus

Never send:

• Asking Jev to generate the answer in the same call
• Dumping the whole corpus because tokens are “cheap” — distractors still hurt jev-1.13
• Raw PDFs / images

Shape the payload like this once the gather step finishes:

{
  "query": "What is the refund window for pro plans?",
  "passage": { "id": "doc-88#p3", "text": "Pro subscribers may request a refund within 14 days." },
  "corpus": { "trust": "internal_kb" }
}

Decision signals and actions

Each evidence field should change a named answer:

TypeSafe’s parallel-questions cookbook: batch questions on one state. For many candidates, loop pairs (query, passage) or shortlist with BM25 first — official rerank cookbook — instead of one giant Choice over 200 ids unless you followed their line-search pattern.

Do not treat a Noul of 0.5 as a “medium” RAG passage decisions score — it means yes and no are equally likely. Conjunctions stay in your code.

Guardrails and escalation

If the gather step fails (empty query + retrieved passage, redaction stripped everything, retriever empty), fail closed on showing a passage to a customer-facing answerer. Do not invent evidence so Jev has something to say. TypeSafe’s confidence-gated examples use a lower bar for recoverable reads than for irreversible actions. Those numbers are illustrations. For RAG passage decisions, treat feed_to_answerer as the high bar (showing a passage to a customer-facing answerer). Tune on labels — see offline evaluation.

Evaluation and rollout notes

Your eval set should include thin-evidence cases, not only happy query + retrieved passages. Label passage relevant / not, plus injection gold on a hostile slice. Pin jev-1.13.0 (the versioned id) after you fit thresholds. jev-latest and the marketing line jev-1.13 can move. Log the response model. TypeSafe’s published list price for jev-1.13 is $0.042 per million input tokens (vendor claim — confirm on the models page); output tokens are free on that same page. Unused distractors still bill as input.

Official Python and JavaScript SDKs read TYPESAFE_API_KEY and retry documented 429/529. This site does not sell, issue, or proxy TypeSafe keys. Use a credential you already have from the console or a documented gateway.

Pack map

FAQ

Should evidence live in the question text? Put facts in state and point instructions at query, passage.text. Criteria stay stable so you can replay.

When do I split calls? TypeSafe’s parallel-questions cookbook: batch questions on one state. For many candidates, loop pairs (query, passage) or shortlist with BM25 first — official rerank cookbook — instead of one giant Choice over 200 ids unless you followed their line-search pattern.

Where is the rest of the RAG pack? Start with RAG input contracts and RAG decision workflow. Cluster hub: Use cases.

Should Jev generate the RAG answer? No. Classify or score passages; another model (or extractive code) writes. That is the cite-vs-generate boundary.

Do we publish rerank lifts? No. TypeSafe’s cookbooks may show measurements — treat those as vendor figures and re-run on your corpus.

What this page does not claim

• No invented top-1 / top-10 lifts.
• Not a vector database.
• Not official TypeSafe.
• Official TypeSafe status, or that jev.pro issues API keys.
• That a schema-constrained answer is automatically factually correct.

Disclaimer

This is an independent unofficial site and is not affiliated with TypeSafe AI; official documentation is available at https://docs.typesafe.ai.

Primary documentation: https://docs.typesafe.ai. Hub: Use cases.