HASH is an agent-based simulation platform originally developed by HASH Labs. It grew out of a vision to make simulation-driven decision making accessible through a web-based IDE backed by a fast runtime, positioned as an open-source alternative to commercial tools like AnyLogic, FlexSim, and Simio.
Components
The legacy platform had three parts, all open source:
- hCore - a browser-based IDE for authoring agent-based models in JavaScript, Python, or Rust
- hEngine - a Rust runtime that executes hCore models at scale
- hCloud - hosted execution for large parameter sweeps
Current state
HASH has pivoted away from simulation as its primary focus. hCore is deprecated and no longer officially supported. hEngine is published as a pre-release technology preview with an unstable codebase. The company has refocused HASH as an all-in-one workspace for structured knowledge and self-building knowledge graphs, with agent-based simulation a longer-term goal to be revisited through generative-AI-driven behaviour definition. Both the legacy hCore IDE and hEngine remain publicly available on GitHub for anyone who wants to run or study them.