[
  {
    "id": "1",
    "input": "Explain the Bifurcation Directive.",
    "output": "The Prime Directive of Adam v23.5 is 'Bifurcation'. This architectural principle strictly separates the codebase into two paths to balance Reliability and Velocity.\n\nPath A: The Product (Reliability)\nLocated in `core/agents/`, this path prioritizes stability. Code must use strict typing (Pydantic), defensive programming, and comprehensive logging.\n\nPath B: The Lab (Velocity)\nLocated in `experimental/` and `research/`, this path allows for rapid experimentation and 'breaking things' to discover new capabilities.",
    "weight": 1.0,
    "tags": ["architecture", "core-directive", "system-1"]
  },
  {
    "id": "2",
    "input": "How does System 2 differ from System 1?",
    "output": "System 1 (The Swarm) operates asynchronously, handling high-volume, low-latency tasks like data fetching and news monitoring. It follows a 'Fire-and-Forget' pattern.\n\nSystem 2 (The Graph) operates synchronously, handling complex reasoning and multi-step workflows. It follows a State Machine pattern, often implemented via LangGraph, and requires structured inputs and outputs with confidence scores.",
    "weight": 0.9,
    "tags": ["architecture", "cognitive-model", "system-2"]
  },
  {
    "id": "3",
    "input": "What is the role of the RiskAssessmentAgent?",
    "output": "The RiskAssessmentAgent is a System 2 component responsible for evaluating financial risks. It does not simply output a single number; it treats risk as a distribution. Key functions include calculating Value at Risk (VaR), assessing creditworthiness, and identifying operational risks.",
    "weight": 0.85,
    "tags": ["agent", "risk", "financial-analysis"]
  },
  {
    "id": "4",
    "input": "Define 'Neuro-Symbolic Cognitive Engine' in the context of Adam.",
    "output": "A 'Neuro-Symbolic Cognitive Engine' combines the probabilistic power of neural networks (LLMs) with the structured reasoning of symbolic AI (rule-based logic, graphs). Adam v23.5 leverages LLMs for understanding context and generating text, while using symbolic systems (MCP, LangGraph) for structured data handling, tool execution, and deterministic logic.",
    "weight": 0.95,
    "tags": ["ai-theory", "neuro-symbolic", "architecture"]
  },
  {
    "id": "5",
    "input": "What is the purpose of the 'Modular Data Manager'?",
    "output": "The 'Modular Data Manager' is a utility class in the Adam showcase environment. It enables the granular, on-demand loading of large datasets (like reports, credit memos, and file indices) instead of loading a monolithic `mock_data.js` file. This improves initial load performance and supports scalability.",
    "weight": 0.7,
    "tags": ["frontend", "optimization", "data-management"]
  },
  {
    "id": "6",
    "input": "Explain the concept of 'Artisanal Data' for SLM training.",
    "output": "'Artisanal Data' refers to small, high-quality, manually curated or verified datasets used for fine-tuning Small Language Models (SLMs). Unlike massive web-scraped datasets, artisanal data is dense with domain-specific knowledge, reasoning traces, and correct formatting, allowing SLMs to punch above their weight class in specific tasks.",
    "weight": 1.0,
    "tags": ["training", "slm", "data-science"]
  },
  {
    "id": "7",
    "input": "What is LoRA and why is it used here?",
    "output": "LoRA (Low-Rank Adaptation) is a parameter-efficient fine-tuning technique. Instead of retraining all model parameters, LoRA injects trainable low-rank decomposition matrices into each layer of the Transformer architecture. This significantly reduces the number of trainable parameters, allowing for efficient adaptation of large models on consumer hardware.",
    "weight": 0.8,
    "tags": ["training", "lora", "machine-learning"]
  },
  {
    "id": "8",
    "input": "Describe the function of the 'Market Mayhem' module.",
    "output": "'Market Mayhem' is a specialized module within Adam designed to simulate and analyze extreme market conditions and crisis scenarios. It generates archives of market reports, newsletters, and stress test results, helping analysts understand potential impacts of black swan events.",
    "weight": 0.75,
    "tags": ["module", "simulation", "market-analysis"]
  },
  {
    "id": "9",
    "input": "What is the significance of the 'Prime Directive'?",
    "output": "The Prime Directive ensures that the development of Adam balances reliability and innovation. By strictly bifurcating the codebase, developers can rapidly prototype in the Lab without jeopardizing the stability of the Product core.",
    "weight": 0.88,
    "tags": ["governance", "development-process"]
  },
  {
    "id": "10",
    "input": "How does the 'Command Palette' function?",
    "output": "The Command Palette (triggered by Ctrl/Cmd+K) provides a keyboard-centric interface for navigating the Adam system. It allows users to quickly search for reports, agents, and modules, enhancing the efficiency of the workflow.",
    "weight": 0.65,
    "tags": ["ux", "navigation", "frontend"]
  }
]