Full Track: Author It All

You chose the sovereign path. Build everything from first principles.

---

What You'll Build

By the end of this track:

Component	Scope
Connections	Multi-zone sensor network
Surfaces	Cleanroom, QA, Mixing, Batch
Warm queries	Complete compliance library
Team	Role-based access for Ops, QA, Engineering
AI integration	Full MCP connection

This is the comprehensive path—everything you need for a production-ready pharma monitoring system.

---

Phase 1: Design Connection Architecture

Step 1.1: Map Your Data Sources

Before creating connections, map what you'll monitor:

Zone	Data Sources	Connection Type
Cleanroom	Temperature, humidity, pressure	IoT Hub
Mixing	Batch sensors, valve states	IoT Hub
QA	Test results, approvals	Database
Utilities	HVAC, water systems	Simulator

Step 1.2: Create Connections

For each data source:

Option A: Using the UI

1. Drag a Connection node from the Node Bank onto your canvas
2. Click to open Inspector and select type (IoT Hub, Database)
3. Configure settings and commit your changes

Option B: Ask Azi

Azi will propose all three connections. Review and approve each.

Step 1.3: Start with Simulator

If you don't have hardware connected yet:

1. Drag a Simulator node from the Node Bank onto your canvas
2. Click to open Inspector and configure variables for temperature, pressure, and batch data
3. Define a JSON template that mirrors your pharma sensor data structure

---

Phase 2: Create Surface Hierarchy

Surfaces organize your data by function. For pharma, a typical hierarchy:

Pharma-Factory (workspace)
├── Cleanroom-Monitoring
│   └── Environmental sensors
├── Batch-Tracking
│   └── Process data
├── QA-Gating
│   └── Test results
└── Utilities
    └── HVAC, water

Step 2.1: Create Cleanroom Surface

Ask Azi:

"Create a surface for cleanroom monitoring that includes temperature, humidity, and differential pressure."

Azi proposes:

• Surface name: Cleanroom-Monitoring
• Connected nodes: Temperature, Humidity, DifferentialPressure
• Description: Environmental monitoring for ISO 7 cleanroom

Review and approve.

Step 2.2: Create Remaining Surfaces

Repeat for each zone:

Surface	Nodes	Purpose
Batch-Tracking	BatchStart, BatchEnd, Ingredients	Process monitoring
QA-Gating	TestResults, Approvals, Holds	Quality control
Utilities	HVAC, Water, Compressed Air	Support systems

---

Phase 3: Build Warm Query Library

A pharma compliance system needs standard queries. Build them with Azi.

Query Library Overview

Query Name	Purpose	Surface
temp-compliance	24h temperature compliance	Cleanroom
batch-history	Full batch event log	Batch-Tracking
pressure-alerts	Differential pressure violations	Cleanroom
qa-pending	Batches awaiting QA approval	QA-Gating
trend-analysis	7-day trends by parameter	All

Step 3.1: Build Temperature Compliance Query

Ask Azi:

"Create a warm query that calculates daily temperature compliance for the cleanroom, flagging any readings outside 18-22°C."

Azi proposes:

TelemetryData
| where Zone == "Cleanroom"
| where Timestamp > ago(24h)
| extend Hour = bin(Timestamp, 1h)
| summarize
    MinTemp = min(Temperature),
    MaxTemp = max(Temperature),
    AvgTemp = avg(Temperature),
    Readings = count()
    by Hour
| extend Compliant = MinTemp >= 18 and MaxTemp <= 22
| order by Hour asc

Review, approve, save.

Step 3.2: Build Batch History Query

Ask Azi:

"Create a query that shows complete history for any batch ID, including all zones and operators."

Step 3.3: Build Alert Queries

Ask Azi:

"Create a query that identifies any pressure readings outside the 0.5-1.5 bar range in the last hour."

---

Phase 4: Establish Team Permissions

Role-based access ensures the right people see the right data.

Step 4.1: Define Roles

Role	Access	Use Case
Admin	Everything	Engineering, IT
Editor	Create and modify queries	Process engineers
Viewer	Run queries, view data	Operators, QA

Step 4.2: Invite Team Members

Option A: Using the UI

1. Go to Team tab
2. Click Invite Member
3. Enter email and select role

Option B: Ask Azi

Step 4.3: Set Surface-Level Permissions

For sensitive surfaces, restrict access beyond workspace roles:

1. Open surface settings
2. Click Permissions
3. Override workspace role for specific users

---

Phase 5: Connect External AI Tools

Complete the system by connecting MCP-native AI.

Step 5.1: Generate MCP Configuration

1. Open the APIs menu in the top bar
2. Select API Keys to generate a JWT token
3. Copy your workspace ID from the workspace settings
4. Use the configuration below with your credentials

Step 5.2: Configure Your AI Client

For Claude Desktop:

{
  "mcpServers": {
    "pharma-factory": {
      "command": "npx",
      "args": [
        "-y",
        "@openindustrial/mcp-server",
        "--workspace", "YOUR_WORKSPACE_ID",
        "--token", "YOUR_TOKEN"
      ]
    }
  }
}

Step 5.3: Test AI Integration

Ask Claude:

"What's the compliance status for the cleanroom over the last 24 hours?"

Claude calls your temp-compliance warm query and returns governed results.

---

What You Built

Component	Count
Connections	4
Surfaces	4
Warm queries	5+
Team members	Configured
MCP integrations	1+

---

Category Metrics

Metric	Your Result
Time to First Governed Query	~1 hour
Total queries built	5+
Surfaces structured	4
Audit coverage	100%
AI connected	MCP-native

---

What Governed AI Looks Like

You didn't just build a monitoring system. You built proof:

Before Control Plane	After Control Plane
AI experiments in silos	AI governed enterprise-wide
No query audit trail	Every query logged with who/when/what
Vendor lock-in	MCP-native portability
Scattered compliance evidence	Single source of governed truth

---

Next Steps

If you want to...	Go to...
Try the Hybrid track	Hybrid Track →
Deep-dive on queries	KQL Basics →
Learn more about MCP	MCP Guide →
See common patterns	Patterns →