Hatim El JazouliCatalog
F101

Ensemble AI forecasting bot

Forecasting. — c. 2025. — 0 tracings.

An ensemble of language-model forecasters, each simulating a different expert persona, that generates judgmental forecasts.

An ensemble forecasting bot built on large language models. Each member simulates a different expert persona, and together they generate judgmental forecasts.

The bot combines the diverse personas into scored, judgmental predictions.

Built with: Python, LiteLLM, OpenRouter, AskNews.

Tracings

Drawn from this record.

None recorded.

Traced from

Records that point here.

  1. R101Context Tree Switching implementationFormal methods

    Both combine many predictors instead of trusting one. CTS mixes context models as data arrives; the bot combines forecasts from different personas.

  2. F102BetomcatForecasting
  3. N108Steeped in ParadoxesNotes

    The VAR argument is the one I care about in forecasting too: random error averages out, systematic bias does not.