Axle Engine Quick Start

Deploy Axle Engine to Azure in 10 minutes.

What is Axle Engine?

Axle Engine is an enterprise-grade ML API that runs in your Azure environment. All your data stays in your infrastructure - perfect for on-premises requirements.

Perfect for: 500+ vehicle fleets with data sovereignty needs
Deployment: Azure Container Apps (your subscription)
Pricing: $0.15/core-hour or annual BYOL

---

Deployment Options

Pay-As-You-Go

✅ No upfront cost
✅ Auto-scaling
✅ Billed through Azure

Cost: ~$432/month (4 cores, 24/7)

Enterprise BYOL

✅ Fixed annual pricing ($24k/year for 1,000 vehicles)
✅ Unlimited deployments
✅ Custom models

---

Azure Marketplace Deployment

Step 1: Deploy from Azure Marketplace

1. Click the button below or go to Azure Marketplace
2. Click "Get It Now"
3. Sign in with your Azure account

Step 2: Configure

Basics:

• Resource Group: Create new or use existing
• App Name: your-company-ml
• Region: Choose closest to your operations

Sizing:

• CPU: 4 cores (recommended)
• Memory: 8 GB

Input Mode:

• REST (default) - HTTP API
• Kafka - Stream processing
• MQTT - IoT devices

Estimated Cost: $432/month (4 cores, 24/7)

Step 3: Deploy

1. Review + Create
2. Wait 3-5 minutes
3. Get your API URL from deployment outputs

---

Quick Test

Health Check

curl https://your-app.azurecontainerapps.io/v1/health

Expected:

{
  "status": "healthy",
  "version": "2.0.0",
  "models_loaded": 18
}

First Prediction

curl -X POST https://your-app.azurecontainerapps.io/v1/predict \
  -H "Content-Type: application/json" \
  -d '{
    "vehicleId": "TRUCK-001",
    "coolant": 95.5,
    "engine_speed": 2400,
    "battery_voltage": 13.8
  }'

Response:

{
  "vehicleId": "TRUCK-001",
  "health_score": 87.5,
  "health_status": "healthy",
  "risk_level": "low",
  "maintenance_alerts": [],
  "recommendations": ["Continue regular maintenance schedule"],
  "processed_at": "2026-01-14T10:00:00Z"
}

---

What's Included

• 18 Pre-trained ML models - Health scoring, anomaly detection, RUL prediction
• Intelligent Schema Mapping - Auto-maps 50+ field name variations
• REST + Streaming APIs - Kafka and MQTT support
• Auto-scaling - 1-10 replicas based on load
• 99.9% uptime SLA - Enterprise reliability

---

Key Features

On-Prem / Private Cloud

Your data never leaves your Azure subscription. ML models are embedded in the container - no external API calls.

Schema Flexibility

Don't have time to change your field names? Our schema processor auto-maps common variations:

• rpm → engine_speed
• coolant_temp → coolant
• odo → obdDistance

Enterprise Security

• SOC 2 compliant
• No telemetry collection
• All HTTPS traffic
• No admin access required

---

Next Steps

• Telemetry Schema - Field reference and auto-mapping
• API Reference - Complete endpoint documentation
• Pricing - Cost comparison

---

Direct Links

Resource	Link
Azure Marketplace	Get It Now →
Documentation	docs.vehk.in
Support	support@vehk.in