Skip to content

Developers

Behavioral models behind an API.

Connect a warehouse or install an SDK, train a model on your own behavioral data, and serve per-customer predictions at runtime. No ML pipeline to build.

request
curl "https://api.moveo.one/v1/predict?userId=user_123" \
  -H "Authorization: Bearer $MOVEO_TOKEN"
200 OK · application/json
{
  "userId": "user_123",
  "predictions": [
    {
      "model": "checkout_completion",
      "label": "likely_to_complete",
      "confidence": 0.87
    },
    {
      "model": "session_dropout",
      "label": "high_dropout_risk",
      "confidence": 0.68
    }
  ]
}
  1. 01 · ConnectWarehouse, analytics or SDK
  2. 02 · ModelTrained on your behavioral history
  3. 03 · PredictGET /predict at runtime
  4. 04 · ActWebhooks, Slack, your code
01MCP server

Let your AI assistant do the install.

Point Claude, Claude Code, Cursor or any MCP client at your workspace. It installs the SDK, verifies events arrive, answers questions about your data, builds cohorts and runs simulations from your editor.

MCP install guide →
terminal
$ uvx moveo-one-mcp login
→ opening browser to approve the connection
→ workspace connected

Personal API tokens cover headless clients such as n8n or CI.

03Sources and actions

Warehouse-first. Product analytics next to it.

Start with a readiness scan on your own data.

30 days of behavioral capture, one model trained on your data and one simulation. Then Pilot from €1,500 / month when you want predictions live.