Donizo - Bathroom Pricing Engine

A smart pricing engine for bathroom renovation projects that generates structured quotes from natural language transcripts.

cd bathroom-pricing-engine
python3 demo_scenarios.py

This repository contains a complete bathroom pricing engine that:

• Parses natural language transcripts to extract project requirements
• Generates structured JSON quotes with detailed cost breakdowns
• Supports city-based pricing (Paris, Marseille, Lyon, etc.)
• Includes confidence scoring for quote accuracy
• Provides comprehensive testing and demo scenarios

bathroom-pricing-engine/
├── pricing_engine.py          # Main engine
├── demo_scenarios.py          # Demo scenarios for testing
├── pricing_logic/
│   ├── material_db.py         # Material costs
│   ├── labor_calc.py          # Labor calculations
│   └── vat_rules.py           # VAT calculations
├── data/
│   ├── materials.json         # Material database
│   └── price_templates.csv    # Pricing templates
├── output/
│   └── sample_quote.json      # Sample output for reference
├── tests/
│   └── test_logic.py         # Test suite
└── DELIVERABLES_SUMMARY.md   # Project summary

python3 demo_scenarios.py

cd tests && python3 test_logic.py

The engine generates comprehensive JSON quotes with:

• Grouped by zone/task: Demolition, Plumbing, Tiling, etc.
• Labor costs: Hours, rates, and total labor cost per task
• Material breakdown: Detailed material costs with quantities
• VAT calculations: Proper French VAT rates
• Confidence scores: Accuracy assessment for each quote

• Transcript Parsing: Extracts area, location, tasks, budget consciousness
• City-based Pricing: Different multipliers for Paris, Marseille, Lyon, etc.
• Budget Detection: Automatic material downgrades for budget-conscious clients
• Comprehensive Testing: 19 test cases covering all functionality
• Production Ready: Error handling and edge cases covered

• Deliverables Summary - Project overview
• Sample Output - Example quote format

• Python 3.8+ - No external dependencies
• JSON - Structured output format
• Modular Architecture - Clean separation of concerns

• Python 3.8+
• No external dependencies required

cd bathroom-pricing-engine

# Run the main engine
python3 pricing_engine.py

# Run demo scenarios for testing
python3 demo_scenarios.py

# Run the test suite
cd tests && python3 test_logic.py

from pricing_engine import BathroomPricingEngine

engine = BathroomPricingEngine()

transcript = """
Client wants to renovate a small 4m² bathroom. They'll remove the old tiles, 
redo the plumbing for the shower, replace the toilet, install a vanity, 
repaint the walls, and lay new ceramic floor tiles. Budget-conscious. 
Located in Marseille.
"""

quote = engine.generate_quote(transcript)
output_file = engine.export_quote_json(quote)

print(f"Total Price: €{quote.total_price:.2f}")
print(f"Confidence: {quote.confidence_score:.1%}")

The engine generates structured JSON quotes:

{
  "quote_id": "BATH-20241201-143022",
  "client_description": "Original transcript text",
  "location": "marseille",
  "total_area": 4.0,
  "total_materials": 1250.50,
  "total_labor": 1800.75,
  "total_vat": 610.25,
  "total_price": 3661.50,
  "confidence_score": 0.85,
  "generated_at": "2024-12-01T14:30:22.123456",
  "assumptions": [
    "Standard bathroom layout for 4.0m²",
    "Location multiplier applied: 1.1x",
    "Prices include basic fixtures and fittings",
    "Does not include design fees or project management",
    "Valid for 30 days from quote date",
    "Budget-friendly materials and labor rates applied"
  ],
  "items": [
    {
      "task_name": "tile_removal",
      "zone": "Demolition",
      "materials": [...],
      "labor_hours": 2.0,
      "labor_rate": 25.0,
      "material_cost": 4.0,
      "labor_cost": 50.0,
      "subtotal": 54.0,
      "vat_rate": 0.20,
      "vat_amount": 10.8,
      "total": 64.8,
      "confidence_score": 0.85,
      "notes": "Includes disposal of old materials"
    }
  ]
}

• Area: Detected from patterns like "4m²", "small bathroom"
• Location: City-based pricing (Paris, Marseille, Lyon, etc.)
• Tasks: Identified from keywords (tile removal, plumbing, tiling, etc.)
• Budget Consciousness: Detected from words like "budget", "cheap", "affordable"

• Database-driven with costs, availability, and quality levels
• Budget-conscious selection of cheaper alternatives
• Location multipliers (Paris: 1.4x, Marseille: 1.1x)

• Trade-specific rates (plumbers, electricians, tilers, etc.)
• Location-based pricing
• Area-based complexity adjustments
• Task-specific hours

• Task-based VAT rates
• Material VAT rates
• French compliance (20% standard, 10% reduced, 5.5% super-reduced)

• Area-based confidence
• Material availability
• Task complexity

• Tile removal and demolition
• Plumbing work (shower, toilet, sink, vanity)
• Tiling (walls and floors)
• Painting and finishing
• Flooring installation
• Electrical work
• Fixture installation

• Paris (1.4x multiplier)
• Marseille (1.1x multiplier)
• Lyon (1.2x multiplier)
• Toulouse (1.15x multiplier)
• Nice (1.25x multiplier)
• Default locations (1.0x multiplier)

• Budget level: Cheaper alternatives
• Standard level: Mid-range quality
• Premium level: High-end materials

# In pricing_logic/material_db.py
self.materials['new_material'] = Material(
    name='New Material',
    cost_per_unit=50.0,
    unit='m²',
    coverage_per_unit=1.0,
    available=True,
    quality_level='standard',
    supplier='Supplier Name'
)

# In pricing_logic/labor_calc.py
self.labor_calc.update_labor_rate('paris', 'plumber', 60.0)

# In pricing_logic/vat_rules.py
self.vat_rules.update_vat_rate('standard', 0.22)

This project is part of the Donizo founding data engineer test case.