Semogram Docs
Getting started

Getting started

Sign in, select a workspace and project, connect data, review and save a pipeline, then run it and verify your first result with evidence.

  1. Sign in or register.
  2. Select a workspace and project.
  3. Connect data, or upload a small dataset.
  4. Create a pipeline, review its proposal, validate configuration, and save a version.
  5. Run it separately and inspect the execution.
  6. Verify output data and evidence before using the result.

Understand the pieces

Core concepts introduces workspace/project scope, plugins and endpoints, pipelines, ontology, evidence, predictions, review and skills. Read it before setup or return to a topic when a resource or lifecycle is unclear.

Continue by task

Configuration preflight does not execute a sample run. Installation does not start automatic syncing. Use a bounded first execution to verify actual behavior.

What to have ready

Choose a small dataset you are allowed to use and one question with a checkable answer. Have the source connection details or a supported upload ready. Begin with a bounded input so you can compare the result with known records.

Your first milestone is a saved pipeline, a completed execution and an output you have inspected. A connected source or a validated draft is progress toward that milestone, not the finished result.