Review Aspect & Sentiment Labeller

Architecture overview

Offline pipelines can fan out many Jev AI tools workers to tag complaint aspects and suspicious review patterns without paying generative output tokens.

Official API schema (field names from docs.typesafe.ai)

Source: TypeSafe HTTP API reference. Question IDs below this table are directory example compositions — you choose them.

Example composition for this Jev AI tools workflow

Sample state and question keys are illustrative compositions for this directory page — not a separate official product API. Primitives remain Choice / Score / Noul.

Feedback: "Product arrived with crushed packaging and scuffed corners. Customer support was slow to respond."

Runnable call examples

Endpoint: https://api.typesafe.ai/v1/systemone (official). Requires your own TYPESAFE_API_KEY.

curl

curl -X POST https://api.typesafe.ai/v1/systemone \
  -H "Authorization: Bearer $TYPESAFE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"state":{"review":"REVIEW_TEXT"},"model":"jev-latest","questions":{"complaint":{"type":"choice","instructions":"Primary complaint category","criteria":{"damaged_packaging":"Packaging","defective_hardware":"Defect","slow_support":"Support","shipping_delay":"Shipping"}},"is_fraud":{"type":"noul","instructions":"Suspicious bot-farm pattern?"}}}'

TypeScript

import { choice, noul, TypeSafeClient } from "@typesafe-ai/sdk";

const client = new TypeSafeClient();
const results = await Promise.all(
  reviews.map((review) =>
    client.systemOne({
      state: { review },
      model: "jev-latest",
      questions: {
        complaint: choice("Primary complaint category", {
          damaged_packaging: "Packaging damage",
          defective_hardware: "Product defect",
          slow_support: "Support delay",
          shipping_delay: "Late shipping",
        }),
        is_fraud: noul("Suspicious bot-farm pattern?"),
      },
    })
  )
);

Python

from typesafe_sdk import Choice, Noul, TypeSafeClient

with TypeSafeClient() as client:
    results = [
        client.system_one(
            state={"review": review},
            model="jev-latest",
            questions={
                "complaint": Choice(
                    instructions="Primary complaint category",
                    criteria={
                        "damaged_packaging": "Packaging damage",
                        "defective_hardware": "Product defect",
                        "slow_support": "Support delay",
                        "shipping_delay": "Late shipping",
                    },
                ),
                "is_fraud": Noul(instructions="Suspicious bot-farm pattern?"),
            },
        )
        for review in reviews
    ]

Latency & cost (source attribution)

Official end-to-end latency range: ~70–500ms; many calls land near ~100ms from US West Coast (source: typesafe.ai / TypeSafe public materials, 2026-09). Official list price: $0.042 / 1M input tokens; output tokens free (source: typesafe.ai, as of 2026-09). Card latency figures are illustrative compositions within that published range — not independent lab measurements by this directory.

• Card illustration on this page: ~86ms (illustrative, within official range — not a lab run by jevaitools.com).
• Official list price: $0.042 / 1M input tokens; output tokens free (source: typesafe.ai, as of 2026-09).

This directory has not published an independent measurement script for this page. To measure yourself: call the official endpoint with your key, record wall-clock p50/p95 andusage.input_tokens, and keep the date of the run.

Comparison with generative LLMs on the same decision task

TypeSafe publishes workflow evaluations where Jev is compared with frontier LLMs on accuracy, cost, and latency (company materials, 2026). Those multipliers are vendor-reported ceilings, not results measured by this directory. Run the same questions through an LLM structured-output adapter on your labeled set before choosing a stack.

Common failure modes

• Criteria that overlap (support vs shipping)
• Multilingual reviews without language-aware state
• Treating Noul as a calibrated fraud probability without labeled evaluation

Other Jev AI Tools

• Agent Step & Action Router
• Zero-Lag Jailbreak Guard
• High-Throughput Ticket Router
• Dynamic UI Component Dispatcher
• Factuality & Hallucination Scorer
• Browser-Use DOM Element Targeter
• Financial News Sentiment Breaker
• CI/CD Flaky Test Triager