> For the complete documentation index, see [llms.txt](https://docs.karini.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.karini.ai/agentic-ai-workshops/energy-and-utilities.md).

# Energy and Utilities

Hands-on labs that build working agents for utilities on Karini AI and AWS, starting with meter-to-cash delayed bill prevention.

Utilities run on high-volume, rules-heavy operations where small exceptions become customer problems. A missed meter read or a device that stopped reporting can stop a bill from generating, and every day it sits unresolved moves it closer to a regulatory exception and a delayed bill letter. Fixing it means checking the billing system, smart meter (AMI) reads, devices, service orders, and account notes, then getting a reviewer to agree before anything changes. That work is done by hand, case by case, against a daily deadline.

These hands-on labs show how agentic AI takes on that work on Karini AI, running on AWS. Each lab builds a working agent workflow end to end, from prompts, tools, and skills to a deployed recipe, using realistic utility data. Each one keeps people in control: agents read data without changing it, a named reviewer makes the decision, and every recommended system action is documented but never executed by the agent.

## The Labs

| Lab                                                                                                                          | What you build                                                                                                                                                                                                     | Where it fits                        | Time         |
| ---------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | ------------------------------------ | ------------ |
| [BillGuard](/agentic-ai-workshops/energy-and-utilities/billguard-build-an-agent-workflow-for-meter-to-cash-delayed-bills.md) | A webhook-triggered workflow where a router sorts each request, one agent investigates a delayed bill (B108) and emails a brief to a reviewer, and another agent turns the reviewer's reply into a resolution card | Meter-to-cash and billing operations | 60 to 75 min |

More Energy and Utilities labs will be added to this section.

## What You Will Learn

* **Event-driven automation:** a workflow that starts from an outside system through a webhook
* **Human in the loop by email:** an agent emails its findings to a reviewer, and a mail connector reads the reply back into the workflow
* **Multi-agent design:** a router that sends new cases and reviewer replies to the right agent
* **Grounded investigation:** read-only database queries and a data catalog, so every finding traces to the data
* **Skills:** packaged instructions, SQL templates, and an SOP that agents read when they need them
* **Right-sized models:** a small, fast model for routing and a larger model for the agents, chosen side by side in Test & Compare
* **Governance:** read-only data access, a single permitted recipient, recommendations that are never executed, and traces that show every step

## Before You Begin

* A Karini AI account at [https://app.karini.ai](https://app.karini.ai/), set up by your Karini AI representative
* A short platform orientation with a Karini AI instructor
* Templates, skills, the workshop database, and credentials prepared by your instructor for each lab
* An email address you can read during the lab, since you act as the reviewer

## Where This Leads

The workshop database stands in for SAP IS-U billing data, the AMI daily reads portal, and the daily reconciliation report. In production, the same agents connect to those systems of record, work the full daily reconciliation queue, and hand approved resolutions to the billing team or a controlled write-back step. The same platform, patterns, and governance carry each agent from a workshop exercise to production.