Configuration

Research Modules

The library ships with several research modules that can be activated and configured independently:

import { Daniel } from '@daniel-murnane/core';

const daniel = new Daniel({
  modules: {
    graphNeuralNetworks: { enabled: true, priority: 'high' },
    particleTracking: { enabled: true, detector: 'ATLAS' },
    mlForPhysics: { enabled: true },
    openScience: { enabled: true }
  }
});

Module: Graph Neural Networks

{
  graphNeuralNetworks: {
    architecture: 'interaction-network',
    application: 'charged-particle-tracking',
    scale: 'HL-LHC',       // High-Luminosity LHC
    performance: 'production-ready'
  }
}

Description:

Key outputs:

Module: ATLAS Tracking & Reconstruction

{
  particleTracking: {
    detector: 'ATLAS',
    subsystem: 'inner-tracker',
    phase: 'HL-LHC-upgrade',
    role: 'ML-co-convener'
  }
}

Description:

Module: Physics-Informed Machine Learning

{
  mlForPhysics: {
    approach: 'physics-informed',
    techniques: ['geometric-deep-learning', 'equivariant-networks', 'graph-networks'],
    domain: 'high-energy-physics'
  }
}

Description:

Module: Open Science & Community

{
  openScience: {
    contributions: ['iris-hep', 'hsf', 'atlas-software'],
    philosophy: 'reproducible-research',
    tools: ['pytorch', 'pytorch-geometric', 'acts']
  }
}

Description:

Runtime Configuration

These settings can be adjusted dynamically:

Option	Type	Default	Description
coffeeLevel	string	'high'	Required caffeine threshold for operation
timezone	string	'Europe/Copenhagen'	Current deployment timezone
mode	string	'research'	Active operating mode
maxConcurrency	number	3	Maximum simultaneous research threads
deepFocusEnabled	boolean	true	Enable uninterruptible focus blocks