Skip to main content

Senzing Cookbook

The Senzing Cookbook

Build a working Senzing solution for fraud or customer-360 by handing plain-English prompts to an AI coding assistant. It does the work: linking the records that refer to the same person or organization across your data sources, then reporting on and exploring them. You don’t write code — you paste prompts and watch it cook.

You’re the chef; the sous-chef is an AI assistant interpreting each prompt in the moment, backed by the kitchen staff — the Senzing MCP, a brigade of specialists. Recipes are starting points; your result won’t be identical to the demo, but it’ll usually be close. Adjust as you go.

Who this is for

A technically comfortable builder — developer, data engineer, analyst, or SE — with a fraud or customer-360 use case. You don’t need to know Senzing or entity resolution (the MCP handles that) and you don’t write code — but you should be comfortable installing dev tools and working through the occasional setup snag. Not a turnkey app for a non-technical user.

Start here — three steps

01

Get set up (once)

line up your AI assistant, the Senzing MCP, and a license. This is the part that takes the most time the first time; the cooking is quick after.

Get Started

02

Pick a recipe

browse the catalog by use case and difficulty.

Browse Recipes

03

Cook

open the recipe and paste each prompt into your assistant, in order.

Good to know

  • → How it works: the cooking metaphor (chef · sous-chef · kitchen staff) lives in How It Works — a one-page visual comparison is metaphor-comparison.html / .pdf.
  • → The one House Rule, baked into every recipe prompt verbatim:
    — Use the Senzing MCP. Do not rely on general training.
  • → Roadmap — what’s under consideration (not commitments).