Decision Models

DECISION AI / IN PRACTICE

From incoming data
to useful decisions.

Experiments in decision AI, with the choices, source evidence and running costs in view.

Explore the showcases
THE APPROACH
  1. 01

    Start with a source.

    Changing information from the real world.

  2. 02

    Make a bounded choice.

    A model selects from defined outcomes.

  3. 03

    Keep it inspectable.

    Follow the evidence and see the cost.

01 / WORKING EXPERIMENTS

Open a showcase.

Real inputs. Visible decisions.

LIVE DEMO / BLUESKY

Conversation radar

Watch topics emerge from a sampled public conversation. Clef screens candidate phrases before they appear; each topic opens its source posts and model decisions.

  • Live public stream
  • Publish / review / reject
  • Sources and model costs
Open the radar How this experiment works
INSIDE THE EXPERIMENT
OBSERVEPublic conversationA bounded ten-minute sample
DECIDE / CLEF-FLASHScreen emerging phrases
publishreviewreject
EXPLORE

Topics, with their evidence.

A view of the process, not live radar data.

02 / THE PROJECT

Build it. Inspect it.
Learn from it.

Decision Models is a TamTa Lab project exploring practical uses of models that choose between defined outcomes.

This is a home for working experiments and writing about what they reveal: where a model helps, how its decisions can be checked, and what it costs to keep it running.

Each showcase has its own purpose and limitations. Start with the original sources, inspect the choices, and draw your own conclusions.

More experiments from TamTa Lab