• TIMING
  • LOCATION
  • IMPACT
  • HUMAN

How to optimize compute workloads with your AI

Connect your AI to live grid data to find the best windows for heavy processing. Your agent identifies when and where the energy mix is cleanest.

Ask AI about this page

Short answer

How can I run compute when carbon intensity is low?

Your AI pulls real-time data to pinpoint energy zones with the lowest emissions. It then suggests the best times to trigger your workloads based on current grid conditions. You get a clear plan for low-carbon scheduling.

The outcomes.

Where your data lands.

The AI turns raw grid telemetry into a scheduling strategy.

TIMING

Optimal run windows

The AI identifies specific hours when carbon intensity drops. You get a list of timestamps for your batch jobs.

LOCATION

Greenest regions

The AI compares different energy zones to find the cleanest spots. You'll know which geographic areas to target for compute.

IMPACT

Emissions visibility

The AI calculates the carbon intensity for your chosen windows. You get a clear view of the footprint you're avoiding.

HUMAN

Final execution

The AI provides the intelligence, but you decide when to actually trigger the jobs. You keep control over your compute clusters.

The workflow

What your AI does when the data arrives.

Your agent handles the heavy lifting of scanning the grid for you.

  1. Scan energy zones

    The AI checks all available regions to see where your compute could run.

    list_energy_zones

  2. Check current intensity

    The agent pulls the real-time carbon footprint for specific locations.

    get_carbon_intensity

  3. Compare regions

    The AI looks across different zones to find the lowest emission spots.

    list_energy_zonesget_carbon_intensity

  4. Build a schedule

    The agent maps out the best times to run tasks based on the data it found.

    get_carbon_intensity

Try it

Copy these to start.

Use these prompts to kick off the optimization process.

Starting points

Swap out the specific regions or task types to match your actual setup.

Electricity Maps Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_5s6xaWAjk6DevLIg1aSrgtFgo5vgeNy7Wwmxo5j7/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — Electricity Maps capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "electricity-maps-carbon-intelligence-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_5s6xaWAjk6DevLIg1aSrgtFgo5vgeNy7Wwmxo5j7/mcp"
    }
  }
}
Copy into chat04

  • Which energy zones currently have the lowest carbon intensity?

  • Find me the cleanest time to run a heavy training job in the US-East zone.

  • Compare the carbon intensity between the UK and Germany energy zones right now.

  • Give me a schedule for when to run my batch processing this week to minimize emissions.

  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Start here

Connect Electricity Maps Carbon Intelligence once, then ask.

One click connects this MCP to your client. Your credentials stay encrypted and your agent gets immediate access to live grid data.

Connect Electricity Maps Carbon Intelligence to your AI

FAQ

How this task behaves.

  • 01

    Can the AI automatically start my compute jobs?

    No. The AI only provides the data and recommendations. You must manually trigger your workloads in your own environment.

  • 02

    Does the AI change my cloud provider settings?

    No. This MCP only reads carbon intensity data. It cannot access or modify your AWS, Azure, or Google Cloud configurations.

  • 03

    How often does the carbon data update?

    The AI pulls real-time data through the connector to ensure you are looking at the most current grid intensity.

  • 04

    Can I use this to track my historical emissions?

    No. The tools are designed to get current and real-time intensity data for planning, not to pull historical logs.

  • 05

    Can the AI tell me which specific data center is cleanest?

    No. The AI can only see data at the energy zone level, not for individual buildings or specific server racks.

  • More questions about Electricity Maps Carbon Intelligence? The Connector page answers them. See everything the Electricity Maps Carbon Intelligence Connector can do

Connect Electricity Maps to Claude, Cursor, ChatGPT & more