JTSense Docs

#Overview

JTSense is a realtime AI intelligence infrastructure designed for crypto markets, AI ecosystems, and onchain monitoring. Built on top of JTVO inference infrastructure, JTSense continuously processes eight live data streams and transforms raw market noise into structured, AI-powered intelligence.

Continuously processed sources

• X (Twitter)
• Onchain transactions
• Wallet activity
• GitHub updates
• Telegram signals
• Narrative trends
• Liquidity movements
• Launch activity

#Core Vision

Crypto markets move faster than humans can process. JTSense creates a realtime AI inference layer that continuously scans and analyzes the market — never sleeping, never blinking. Where a human analyst sees fragments, JTSense fuses every stream into a single coherent picture and surfaces what matters.

#Quickstart

Get from zero to your first signal in under a minute. The full walkthrough lives in the Getting Started guide.

# 1 · authenticate (demo credentials work in the live app)
curl https://api.jtsense.ai/v1/auth \
  -d '{"email":"demo@jtsense.ai","password":"demo"}'

# 2 · open a realtime signal stream
wscat -c wss://stream.jtsense.ai/v1/signals?token=$JWT

# 3 · signals arrive as structured JSON, classified by JTVO
< {"type":"NARRATIVE","label":"AI Agents","confidence":0.92}

#The Seven Engines

Each engine specializes in a slice of the market. Together they form one complete realtime picture, routed through JTVO inference.

1 · Narrative Intelligence Engine

• Detect emerging narratives
• Ecosystem momentum tracking
• Social velocity analysis
• Attention rotation detection

2 · Wallet Intelligence

• Smart money accumulation
• Insider wallet tracking
• Whale movement detection
• Liquidity migration analysis

3 · Social Intelligence Layer

• X monitoring
• Telegram analysis
• Sentiment classification
• Coordinated marketing detection

4 · GitHub Intelligence

• Commit tracking
• Repository growth
• Contributor analysis
• Infrastructure update detection

5 · Launch Detection System

• New token deployments
• AI ecosystem launches
• Launchpad monitoring
• Suspicious deployer detection

6 · AI Risk Engine

• Rug pattern detection
• Liquidity analysis
• Engagement farming detection
• Wallet concentration analysis

7 · Realtime Alert System

• Dashboard alerts
• Telegram alerts
• Webhook integrations
• Realtime anomaly notifications

#Data Sources

JTSense ingests from six classes of upstream providers:

#Processing Pipeline

Eight stages turn raw events into delivered intelligence:

1  Data ingestion        → pull from all upstream sources
2  Event normalization   → unify into schema.v3
3  Signal extraction     → anomalies, velocities, patterns
4  JTVO inference routing → dispatch to inference layer
5  AI classification     → type + confidence scoring
6  Intelligence synthesis→ cross-source reasoning
7  Alert generation      → threshold + rule evaluation
8  Realtime delivery     → dashboard · telegram · webhook

#Suggested Tech Stack

Backend

• Node.js / Bun
• PostgreSQL · Redis
• Kafka / RabbitMQ
• Docker · Kubernetes

Frontend

• Next.js
• TailwindCSS
• WebSocket realtime feeds
• Terminal-style UI

#API Reference

All endpoints are versioned under /v1 and require a Bearer JWT obtained from the auth endpoint.

Exchange credentials for a JWT.

List recent signals. Query: type, min_confidence, limit.

Ranked narratives with momentum + social velocity.

Wallet profile: label, smart-money score, recent flow.

Realtime signal stream. Authenticate with ?token=$JWT.

GET /v1/signals?type=WHALE&min_confidence=0.8&limit=2
Authorization: Bearer $JWT

{
  "data": [
    { "id": 1246, "type": "WHALE", "label": "Large ETH Accumulation",
      "confidence": 0.89, "source": "onchain", "ts": "2026-05-31T09:14:22Z" }
  ],
  "meta": { "count": 1, "model": "jtvo-infer-l" }
}

#Alerts & Webhooks

Alerts fire when a signal crosses your configured rules. Delivery channels:

• Dashboard — realtime in-app notifications
• Telegram — push to a channel or DM
• Webhook — HTTP POST to your endpoint

Webhook payload

POST https://your-server.com/jtsense-hook
{
  "event": "signal.alert",
  "rule": "whale_accumulation > 100 ETH",
  "signal": {
    "type": "WHALE",
    "label": "Large ETH Accumulation",
    "confidence": 0.89,
    "wallet": "0x7f3a…e21",
    "delta": "+482 ETH"
  },
  "ts": "2026-05-31T09:14:22Z"
}

#Signal Schema

#Branding

Name

JTSense

Slogans

• The realtime intelligence layer for crypto.
• AI-powered market intelligence.
• Realtime inference. Realtime signals.
• Built on JTVO. Powered by JTVO Inference.

Visual Direction

• Terminal UI
• Cyberpunk style
• Dark dashboards
• AI infrastructure aesthetic