Skip to content

Scelo IDE

Scelo is a desktop workbench for actuaries. It takes you from raw data to board-pack-ready results in one place, with a bundled Python and R runtime, the actuarial stack (lifelib, chainladder, climada, and more), and AI co-pilots scoped to each stage of the work. It runs fully offline once installed.

The Scelo pipeline — Soft Data → Tools → Hard Data

  • Soft Data

    Load a CSV / Parquet, inspect every column, clean and reshape it by click or by chat.

  • Tools

    Pick the right actuarial models, or let Scelo suggest them for your dataset's shape and domain.

  • Hard Data

    Run the models, read the results on a canvas, and export a printable board-pack PDF.

  • The swarm

    Convene a multi-agent council to pressure-test a forecast, and simulate a population's response to a scenario.

What makes Scelo different

  • Offline + private. The bundled Python/R runtime and a local LLM (Ollama) mean your client data never has to leave the machine. Hosted AI providers are opt-in.
  • A pipeline, not a pile of scripts. Soft data → tools → hard data is a guided flow where each stage carries its own scoped AI assistant.
  • Two surfaces in one app. A guided pipeline for the analysis, and a full VS Code-style workspace (editor, terminal, Git, search) for when you want to drop into code.
  • Reproducible. Every action you take can be exported as a runnable Python, R, or C++ script.

How this manual is organised

Section What's in it
Installation Get Scelo onto Linux, Windows, or macOS
Getting started Your first run, end to end
The workspace Editor, terminal, Git, search, command palette
The pipeline Soft Data, Tools, Hard Data in depth
The swarm Council, society pulse, and simulation
AI providers Ollama and hosted providers
Chat The scoped assistants throughout the app
Exporting Code export and the board-pack PDF
Reference Shortcuts, file locations, troubleshooting

New here?

Install Scelo (Linux · Windows/macOS), then follow Getting started for a full walk-through.