<<<<<<< HEAD

Corporate Event Intelligence Engine

Evidence-based historical analog retrieval and analyst-style synthesis for corporate events.

This repository is a portfolio-ready analytics product demo focused on CEO departure events. It uses a curated historical case library, deterministic retrieval logic, pathway aggregation, and a static dashboard to show how corporate event intelligence can be structured without forecasting, stock prediction, or investment recommendations.

Product Boundary

• Historical analogs only
• Descriptive analysis only
• No forecasts
• No stock prediction
• No investment advice
• No LLM APIs
• No new event categories in the current demo

Current Demo Question

What usually happens after an abrupt CEO departure?

The demo retrieves comparable CEO departure cases, explains why they were selected, aggregates observed pathways, links the evidence base, and renders an analyst-style brief.

Architecture

flowchart TD
    A["CEO Departure Case Library<br/>data/corporate_events_seed.csv"] --> B["Deterministic Analog Engine<br/>src/historical_analog_engine.py"]
    C["Event Taxonomy<br/>data/event_taxonomy.yaml"] --> B
    D["Pathway Labels<br/>data/pathway_labels.yaml"] --> B
    B --> E["Retrieval Results<br/>outputs/sample_retrieval_results.json"]
    B --> F["Analyst Brief<br/>outputs/sample_ceo_departure_answer.md"]
    E --> G["Static Dashboard<br/>dashboard/index.html"]
    F --> G

Loading

Repository Structure

data/
  corporate_events_seed.csv
  event_taxonomy.yaml
  pathway_labels.yaml
docs/
  product_brief.md
  mvp_architecture.md
  historical_analog_engine_design.md
  retrieval_scoring_framework.md
  sprint2_implementation_plan.md
outputs/
  sample_retrieval_results.json
  sample_ceo_departure_answer.md
src/
  historical_analog_engine.py
dashboard/
  index.html
  styles.css
  app.js

Demo Workflow

Regenerate the deterministic retrieval outputs:

python3 src/historical_analog_engine.py

Validate the engine:

python3 -m py_compile src/historical_analog_engine.py

Serve the dashboard from the repository root:

python3 -m http.server 8000

Open:

http://localhost:8000/dashboard/

Dashboard Screenshot

Placeholder:

docs/assets/dashboard-screenshot-placeholder.svg

Add a screenshot after opening the local dashboard and confirming the JSON and analyst brief render correctly.

Retrieval Logic

The historical analog engine scores cases using:

• departure_type
• sector
• context
• observed_pathway

For the current demo question, sector is neutral because the question does not specify an industry constraint. The output is deterministic: the same case library and query profile produce the same retrieval JSON and analyst brief.

Validation Checklist

• Run python3 src/historical_analog_engine.py
• Run python3 -m py_compile src/historical_analog_engine.py
• Serve the dashboard locally from the repository root
• Confirm outputs/sample_retrieval_results.json renders in the dashboard
• Confirm outputs/sample_ceo_departure_answer.md renders in the dashboard
• Confirm browser console has no errors
• Confirm README links point to existing repository files

Limitations

This product summarizes historical cases and observed pathways. It does not predict future outcomes, recommend investments, issue ratings, or produce price targets.

corporate_event_intelligence_engine

Evidence-based corporate event intelligence engine using historical analogs, observed pathways, and analyst-style synthesis.

c9ae978f832fb8f5da2e47904edf11411c3cb734