PHARMA FACTORY - FULL

Full Track: Author It All

You chose to build every part. Start from first principles.

This track assumes nothing: no legacy systems, no inherited constraints. You design the connection architecture, build the surface hierarchy, and author every warm query.

What you will build

Component	Scope
Connections	A multi-zone sensor network
Surfaces	Cleanroom, QA, mixing, batch
Warm queries	A compliance query library
Team	Access for operations, QA and engineering
MCP	Any AI client reading the library

Phase 1: Design the connection architecture

Step 1.1: Map your data sources

Zone	Data sources	How it reaches OpenX
Cleanroom	Temperature, humidity, pressure	A connection
Mixing	Batch sensors, valve states	A connection
QA	Test results, approvals	An API Source polling your LIMS, or a CosmosDB container a warm query reads
Utilities	HVAC, water	A Simulator until the meters are wired

Step 1.2: Create connections

On the canvas

1. Drag a Connection node from the Node Bank onto the canvas for each zone.
2. In each inspector's Settings tab, set the Connection ID. There is no connection string to enter.
3. Save, then Deploy. Each connection's Connection Info tab then shows the credentials its devices send with.

Or ask Azi

Step 1.3: Start with a simulator

If hardware is not connected yet:

1. Drag a Simulator node from the Node Bank onto the canvas.
2. In its inspector, add Variables for temperature, pressure and batch data.
3. In Template, write a JSON message that mirrors your sensors; $.Name pulls in a variable.
4. Save, then Deploy starts it.

Phase 2: Create the 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: Cleanroom surface

"Create a surface for cleanroom monitoring, fed by the cleanroom sensors connection."

Azi proposes a Cleanroom-Monitoring surface wired to it. Accept it; it is live at once.

Step 2.2: The other surfaces

Surface	Fed by	Purpose
Batch-Tracking	Mixing batch connection	Process monitoring
QA-Gating	A connection that reads the LIMS API source	Quality control
Utilities	The utilities simulator	Support systems

Phase 3: Build the warm query library

Query	Purpose	Surface
temp-compliance	Hourly temperature compliance	Cleanroom
batch-history	Every event for one batch	Batch-Tracking
pressure-alerts	Pressure outside its range	Cleanroom
qa-pending	Batches awaiting QA	QA-Gating
trend-analysis	7-day trends by parameter	All

Step 3.1: Temperature compliance by hour

"Create a warm query for hourly cleanroom temperature compliance over the last day, flagging any hour outside 18-22°C."

Data
| where EnqueuedTime > ago(24h)
| extend Payload = todynamic(RawData)
| where tostring(Payload.zone) == "Cleanroom"
| summarize MinTemp = min(todouble(Payload.temperature)),
            MaxTemp = max(todouble(Payload.temperature)),
            AvgTemp = avg(todouble(Payload.temperature)),
            Readings = count()
            by Hour = bin(EnqueuedTime, 1h)
| extend Compliant = MinTemp >= 18 and MaxTemp <= 22
| order by Hour asc

Step 3.2: Batch history

"Create a query that shows the complete history of a batch, across every zone."

Step 3.3: Pressure alerts

"Create a query for pressure readings outside 0.5-1.5 bar in the last hour."

Phase 4: Give your team access

The owner adds teammates in Workspace -> Team Members: enter their email and click Invite. They are members at once. No AI can invite members, and the role picker in that window is not saved - what a member can change is set by access rights.

Team	What they need
Engineering, IT	Workspace.Deploy to build connections; Workspace.Access.Manage to manage access
Process engineers, QA	Membership: any member can save and accept queries; some changes need a management right (see Permissions)
Operators	Membership, to read the surfaces and run the queries

Access rights apply to the whole workspace; there is no per-surface permission. See Permissions.

Phase 5: Connect any AI over MCP

The OpenX MCP Server is at https://openx.fathym.com/ox-api/mcp. Sign in through the browser: in Claude.ai or ChatGPT, add a custom connector and paste the URL; for coding agents, run fai mcp install https://openx.fathym.com/ox-api/mcp --name openx and sign in when the browser opens. Setup for each client is in Bring Any AI.

Then ask:

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

The AI calls your temp-compliance query with call_api and answers from it. If it wants to build something new, it proposes it with manage_proposals, and a person accepts.

What you built

Component	Count
Data sources	2 connections, an API source and a simulator
Surfaces	4
Warm queries	5
Team access	Set
MCP clients	1 or more

What you can show

Before	After
AI experiments in silos	Every AI client reads the same saved queries
Answers nobody can explain	The query behind each answer, there to read
Scattered compliance evidence	One workspace, organized by surface

What OpenX records about changes, and how it maps to data-integrity requirements, is on the Compliance matrix.

Next steps

If you want to...	Go to...
Try the hybrid track	Hybrid track
Learn KQL	KQL Basics
Connect more AI clients	Bring Any AI
See query patterns	Query Patterns