> For the complete documentation index, see [llms.txt](https://docs.akenza.io/akenza.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.akenza.io/akenza.io/genio/genio.md).

# Overview

Genio is the AI assistant built into akenza. Ask questions about your IoT data in plain language and get answers based on your own devices.

**Genio** is the AI assistant built into akenza. Instead of building a dashboard for every question you might want to ask, you ask the question directly, in plain language, and Genio answers it from the data your devices have already sent.

{% hint style="info" %}
Genio is currently in beta and available at the organization level. See [Applications](/akenza.io/general/applications.md) for how the akenza applications relate to each other, and [akenza.io/genio](https://akenza.io/genio) for access.
{% endhint %}

## What you can ask

Genio answers questions about the devices, data, and deployments in your organization. Typical questions fall into a few categories:

<table><thead><tr><th width="220">Kind of question</th><th>Examples</th></tr></thead><tbody><tr><td><strong>Real-time state</strong></td><td>"How many people are in the office right now?"<br>"How many desks are currently occupied?"</td></tr><tr><td><strong>Historical &#x26; aggregated</strong></td><td>"What was the highest measured temperature in the office this month?"<br>"How did energy consumption develop over the last quarter?"</td></tr><tr><td><strong>Fleet health</strong></td><td>"Which devices are offline?"<br>"Which sensors need a battery replacement soon?"</td></tr><tr><td><strong>General knowledge</strong></td><td>"What is the difference between OTAA and ABP?"<br>"What does a low SNR value tell me about my LoRaWAN device?"</td></tr></tbody></table>

Because Genio works from your own data rather than from a fixed set of reports, the same setup serves different people differently. A technician can look for patterns in equipment downtime, a facility manager can check occupancy and comfort, and a business owner can ask about cost and efficiency, without anyone needing to build a dashboard or write a query first.

## Use Genio to generate detailed reports

Genio is not limited to single-value answers. It can assemble a report that combines several metrics over a period, applies your own cost assumptions, and ends with a recommendation. Space optimization is a good example, because it needs occupancy history, capacity, and cost inputs at the same time.

A prompt for such a report might look like this:

> Build a space optimization report for the 2nd floor covering the last quarter. Compare concurrent and time-based occupancy per zone, identify how many desks were never needed at the same time, and calculate the annual saving and return on investment if we reduce the desk count accordingly. Assume a fully loaded cost of CHF 3,600 per desk per year and a one-off deployment cost of CHF 42,000.

Genio then works through the same reasoning an analyst would:

{% stepper %}
{% step %}

### Establish the utilization baseline

Genio reads the occupancy history per zone and reports both **concurrent occupancy**, the peak number of desks in use at the same time, and **time-based occupancy**, the share of time desks were in use. The gap between the two is where optimization potential hides. See [Occupancy monitoring](/akenza.io/analytics-use-cases/occupancy-monitoring.md) for what each metric measures and when to trust which.
{% endstep %}

{% step %}

### Identify releasable capacity

Desks that never contributed to a peak are candidates for release. Genio checks for [capacity crunch](/akenza.io/analytics-use-cases/occupancy-monitoring.md) periods first, so that a recommendation never cuts into demand that actually occurred.
{% endstep %}

{% step %}

### Apply your cost model

Genio has no knowledge of your rent, cleaning contracts, or utility rates, so you supply them in the prompt or as [custom fields](/akenza.io/get-started/your-organization/how-to-use-custom-fields-on-akenza.md) on the relevant assets. It multiplies the releasable capacity by the cost per desk to get a gross annual saving.
{% endstep %}

{% step %}

### Calculate the return on investment

Set against the cost of the sensor deployment, the saving becomes a payback period and a first-year return.
{% endstep %}
{% endstepper %}

The resulting report reads roughly as follows:

<table><thead><tr><th width="330">Metric</th><th>Value</th></tr></thead><tbody><tr><td>Capacity, 2nd floor</td><td>120 desks</td></tr><tr><td>Peak concurrent occupancy</td><td>78 desks</td></tr><tr><td>Time-based occupancy</td><td>46 %</td></tr><tr><td>Desks that can be released</td><td>30</td></tr><tr><td>Fully loaded cost per desk per year</td><td>CHF 3,600</td></tr><tr><td><strong>Gross annual saving</strong></td><td><strong>CHF 108,000</strong></td></tr><tr><td>Deployment cost, one-off</td><td>CHF 42,000</td></tr><tr><td><strong>Net saving, first year</strong></td><td><strong>CHF 66,000</strong></td></tr><tr><td>Payback period</td><td>4.7 months</td></tr><tr><td>Return on investment, first year</td><td>157 %</td></tr></tbody></table>

Reducing 120 desks to 90 still leaves headroom above the observed peak of 78, so the saving does not come at the cost of the workplace experience. You can keep going from there in the same conversation, asking Genio to break the figures down per zone, to exclude holiday weeks, or to model a more conservative scenario.

{% hint style="warning" %}
The figures above are illustrative. The quality of any cost or ROI report depends entirely on the assumptions you provide and on the occupancy data being complete for the period in question. Treat the output as a basis for a decision, not as the decision itself.
{% endhint %}

Reports of this kind are not limited to space optimization. The same pattern applies to energy savings after a retrofit, maintenance cost avoided through early battery replacement, or the cleaning effort saved by switching from a fixed schedule to occupancy-driven cleaning.

## Use Genio to troubleshoot your fleet

Genio is also part of the **workspace home**, where it works on the condition of your deployment rather than on the measurements it produces. Alongside the workspace **health score** and the uptime metrics of your fleet, Genio helps you manage, debug, and troubleshoot the health of an IoT deployment: why the health score moved, which devices stopped reporting and when, whether a group of devices shares a root cause such as one failing gateway or one broken decoder, and what to do about it.

This makes the workspace home a practical starting point for day-to-day fleet operations. You see the state of the deployment, and you can ask about anything that looks wrong without leaving the page.

{% hint style="info" %}
The redesigned organization and workspace home, including the health score and Genio assistance, was introduced in v2.60.0. See the [Changelog](/akenza.io/changelog.md).
{% endhint %}

## Capabilities

* **Natural language queries.** Ask open-ended questions without technical expertise or query syntax.
* **Real-time and historical analysis.** Genio reads both the latest values and the stored history of your devices to identify trends and patterns.
* **External data sources.** Enrich your IoT data with third-party inputs such as weather APIs, so answers account for context your sensors do not measure.
* **Fleet troubleshooting.** From the organization and workspace home, Genio helps you interpret the health score and uptime metrics of a deployment and suggests how to improve its condition.
* **Contextualized data.** Where the [Data Fusion Layer](/akenza.io/data-fusion-layer/data-fusion-layer.md) is in use, Genio receives telemetry already normalized into a spatial model of sites, floors, zones, and assets, rather than as isolated device readings.

## Where Genio gets its data

Genio does not read your devices directly. It queries what your [data flows](/akenza.io/get-started/your-data-flow.md) have already decoded and stored, along with device metadata such as [custom fields](/akenza.io/get-started/your-organization/how-to-use-custom-fields-on-akenza.md) and [tags](/akenza.io/get-started/your-organization/how-to-use-tags-on-akenza.md).

{% hint style="warning" %}
Genio can only answer questions about data that is actually stored. A data flow that forwards uplinks to an external destination without also writing to the [akenza DB](/akenza.io/get-started/your-data-flow/connectors/databases/akenza-db.md) connector leaves Genio nothing to query.
{% endhint %}

Metadata is therefore worth maintaining: descriptive names, consistent tags, and populated custom fields give Genio the vocabulary to resolve what "the meeting rooms on the second floor" refers to. But metadata describes a device, not the building it sits in, and most useful questions are really questions about the building.

The [Data Fusion Layer](/akenza.io/data-fusion-layer/data-fusion-layer.md) closes that gap. It maintains a semantic model of the site, built on a tailored subset of the [Brick Schema](https://brickschema.org/) ontology, which are fused with the sensor readings to supply:

* **Spatial hierarchy.** Which zones belong to which floor, and which floors to which site, so that "the second floor" is a resolvable entity rather than a naming convention.
* **Asset relationships.** Which sensors and BMS systems serve a given room, so that a question about a room gathers every relevant data point.
* **Type and purpose.** What a space is for, such as a meeting room or a plant room, and what an asset measures.
* **Normalized readings.** Payloads from different vendors in one structure, so that mixed hardware can be compared without per-device special cases.

A question like "which meeting rooms on the second floor were too warm during working hours last week" needs all four before the time filter and the threshold can be applied. The same model covers BMS data arriving through [Akenza Building Edge](/akenza.io/akenza-building-edge/akenza-building-edge.md), so one answer can draw on wired plant data and wireless sensors together.

## Data privacy

Genio runs on the GPT-5 series provided through **Azure OpenAI**, inside the secure Azure environment and independently of the public OpenAI service. **Your data is not used to train models.**

## Use cases

Genio is used across scenarios that combine device data with an operational question, including:

* Occupancy monitoring, meeting room usage, and space optimization. See [Occupancy monitoring](/akenza.io/analytics-use-cases/occupancy-monitoring.md).
* Energy consumption analysis
* Building security and motion monitoring
* Health and safety, such as CO₂ levels and workplace comfort
* Maintenance planning, such as identifying devices due for a battery replacement
* Device lifecycle management and fleet health across a deployment


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.akenza.io/akenza.io/genio/genio.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.