Chief Data Officer
                                 |
                    Data Science Centre of Excellence
                                 |
        ┌────────────┬────────────┼────────────┬────────────┐
        │            │            │            │            │
   Risk Data      Credit Data   Marketing     Operations   Innovation
      Science       Science       Data          Data         Lab
       Team           Team        Science        Science      Team
        │            │            │            │            │
        └────────────┴────────────┴────────────┴────────────┘
                                 |
                    Cross-Functional Communities of Practice

Chief Data Officer / Head of Data Science
├── Director of Data Science (Modelling)
│   ├── Credit Risk Modelling Team
│   ├── Fraud Detection Team
│   ├── NLP/AI Team
│   └── Marketing Analytics Team
├── Director of Data Engineering
│   ├── Data Pipeline Team
│   ├── Data Warehouse/Lake Team
│   └── Data Integration Team
├── Director of MLOps
│   ├── Model Deployment Team
│   ├── Model Monitoring Team
│   └── Infrastructure Team
├── Director of Data Governance
│   ├── Data Quality Team
│   ├── Data Privacy/Compliance Team
│   └── Metadata Management Team
└── Head of Product (Data)
    ├── Product Managers (by business unit)
    └── UX/Design Team

Data Analyst → Senior Data Analyst → Data Scientist → Senior Data Scientist → Principal Data Scientist → Director of Data Science

Data Engineer → Senior Data Engineer → Lead Data Engineer → Principal Data Engineer → Director of Data Engineering

ML Engineer → Senior ML Engineer → Lead ML Engineer → Principal ML Engineer → Director of MLOps

Product Manager (Data) → Senior Product Manager → Director of Product (Data) → Head of Data Products

# ===================================================================
# MODULE 8, LESSON 1: BUILDING A FINANCIAL DATA SCIENCE TEAM
# ===================================================================

import matplotlib.pyplot as plt
import networkx as nx
import pandas as pd
import numpy as np
from matplotlib.patches import Rectangle, FancyBboxPatch
import warnings
warnings.filterwarnings('ignore')

# Set style
plt.style.use('seaborn-v0_8-whitegrid')

print("="*70)
print("BUILDING A FINANCIAL DATA SCIENCE TEAM")
print("="*70)

# ----------------------------------------------------------------
# PART A: TEAM STRUCTURE VISUALISATION
# ----------------------------------------------------------------

print("\n" + "-"*60)
print("PART A: Data Science Team Structure")
print("-"*60)

def create_team_org_chart():
    """Create an organisational chart for a data science team."""
    fig, ax = plt.subplots(figsize=(16, 10))
    ax.set_xlim(0, 14)
    ax.set_ylim(0, 10)
    ax.axis('off')
    
    # Define positions for each role (x, y)
    positions = {
        'Chief Data Officer': (7, 9),
        'Director DS': (3.5, 7),
        'Director DE': (7, 7),
        'Director MLOps': (10.5, 7),
        'Director Governance': (1.5, 5.5),
        'Head Product': (12.5, 5.5),
        'Credit Risk': (1.5, 3.5),
        'Fraud Detection': (3.5, 3.5),
        'NLP/AI': (5.5, 3.5),
        'Marketing Analytics': (7.5, 3.5),
        'Data Pipeline': (10.5, 3.5),
        'Data Warehouse': (12.5, 3.5),
        'Model Deployment': (1.5, 1.5),
        'Monitoring': (3.5, 1.5),
        'Infrastructure': (5.5, 1.5),
        'Data Quality': (9.5, 1.5),
        'Privacy': (11.5, 1.5),
        'Metadata': (13.5, 1.5),
        'Product Mgmt': (7.5, 1.5),
    }
    
    # Define roles and their attributes
    roles = {
        'Chief Data Officer': {'color': '#1f77b4', 'size': 800, 'level': 0},
        'Director DS': {'color': '#ff7f0e', 'size': 600, 'level': 1},
        'Director DE': {'color': '#2ca02c', 'size': 600, 'level': 1},
        'Director MLOps': {'color': '#d62728', 'size': 600, 'level': 1},
        'Director Governance': {'color': '#9467bd', 'size': 500, 'level': 2},
        'Head Product': {'color': '#8c564b', 'size': 500, 'level': 2},
        'Credit Risk': {'color': '#e377c2', 'size': 400, 'level': 3},
        'Fraud Detection': {'color': '#e377c2', 'size': 400, 'level': 3},
        'NLP/AI': {'color': '#e377c2', 'size': 400, 'level': 3},
        'Marketing Analytics': {'color': '#e377c2', 'size': 400, 'level': 3},
        'Data Pipeline': {'color': '#7f7f7f', 'size': 400, 'level': 3},
        'Data Warehouse': {'color': '#7f7f7f', 'size': 400, 'level': 3},
        'Model Deployment': {'color': '#bcbd22', 'size': 400, 'level': 3},
        'Monitoring': {'color': '#bcbd22', 'size': 400, 'level': 3},
        'Infrastructure': {'color': '#bcbd22', 'size': 400, 'level': 3},
        'Data Quality': {'color': '#9467bd', 'size': 400, 'level': 3},
        'Privacy': {'color': '#9467bd', 'size': 400, 'level': 3},
        'Metadata': {'color': '#9467bd', 'size': 400, 'level': 3},
        'Product Mgmt': {'color': '#8c564b', 'size': 400, 'level': 3},
    }
    
    # Draw connecting lines
    connections = [
        ('Chief Data Officer', 'Director DS'),
        ('Chief Data Officer', 'Director DE'),
        ('Chief Data Officer', 'Director MLOps'),
        ('Chief Data Officer', 'Director Governance'),
        ('Chief Data Officer', 'Head Product'),
        ('Director DS', 'Credit Risk'),
        ('Director DS', 'Fraud Detection'),
        ('Director DS', 'NLP/AI'),
        ('Director DS', 'Marketing Analytics'),
        ('Director DE', 'Data Pipeline'),
        ('Director DE', 'Data Warehouse'),
        ('Director MLOps', 'Model Deployment'),
        ('Director MLOps', 'Monitoring'),
        ('Director MLOps', 'Infrastructure'),
        ('Director Governance', 'Data Quality'),
        ('Director Governance', 'Privacy'),
        ('Director Governance', 'Metadata'),
    ]
    
    # Draw lines
    for parent, child in connections:
        p_pos = positions[parent]
        c_pos = positions[child]
        ax.plot([p_pos[0], c_pos[0]], [p_pos[1] - 0.2, c_pos[1] + 0.2], 
                'k-', linewidth=1.5, alpha=0.5)
    
    # Draw nodes
    for role, pos in positions.items():
        info = roles.get(role, {'color': 'gray', 'size': 400})
        color = info['color']
        size = info['size']
        
        # Draw rectangle or circle based on level
        if info['level'] == 0:
            # CEO - large rectangle
            rect = FancyBboxPatch((pos[0]-1.2, pos[1]-0.4), 2.4, 0.8,
                                  boxstyle="round,pad=0.1", 
                                  facecolor=color, edgecolor='black', alpha=0.8)
            ax.add_patch(rect)
            ax.text(pos[0], pos[1], role, ha='center', va='center', 
                   fontsize=9, fontweight='bold', color='white')
        elif info['level'] == 1:
            # Directors - medium rectangle
            rect = FancyBboxPatch((pos[0]-0.9, pos[1]-0.35), 1.8, 0.7,
                                  boxstyle="round,pad=0.1", 
                                  facecolor=color, edgecolor='black', alpha=0.8)
            ax.add_patch(rect)
            ax.text(pos[0], pos[1], role, ha='center', va='center', 
                   fontsize=8, fontweight='bold', color='white')
        elif info['level'] == 2:
            # Senior Managers - circle
            circle = plt.Circle(pos, 0.4, color=color, alpha=0.8, edgecolor='black')
            ax.add_patch(circle)
            ax.text(pos[0], pos[1], role, ha='center', va='center', 
                   fontsize=7, fontweight='bold', color='white')
        else:
            # Teams - small rectangle
            rect = FancyBboxPatch((pos[0]-0.6, pos[1]-0.25), 1.2, 0.5,
                                  boxstyle="round,pad=0.05", 
                                  facecolor=color, edgecolor='black', alpha=0.7)
            ax.add_patch(rect)
            ax.text(pos[0], pos[1], role, ha='center', va='center', 
                   fontsize=6, color='white', fontweight='bold')
    
    # Add title
    ax.text(7, 9.8, 'Financial Data Science Team Structure', 
            ha='center', fontsize=16, fontweight='bold')
    ax.text(7, 9.5, 'Hybrid Model', ha='center', fontsize=12, style='italic')
    
    plt.tight_layout()
    plt.savefig('data_science_team_structure.png', dpi=300, bbox_inches='tight')
    plt.show()

create_team_org_chart()
print("Team structure visualisation saved as 'data_science_team_structure.png'")

# ----------------------------------------------------------------
# PART B: SKILLS RADAR CHART
# ----------------------------------------------------------------

print("\n" + "-"*60)
print("PART B: Skills Radar Chart for Data Science Roles")
print("-"*60)

def create_skills_radar():
    """Create a radar chart comparing skills across roles."""
    from math import pi
    
    # Define skills and role proficiencies (0-10)
    skills = ['SQL', 'Python', 'Statistics', 'Machine Learning', 
              'Deep Learning', 'Data Engineering', 'Cloud', 'Data Viz', 
              'Finance Domain', 'Regulatory', 'Communication']
    
    roles = {
        'Data Scientist': [7, 9, 8, 9, 7, 5, 6, 7, 7, 6, 7],
        'ML Engineer': [7, 9, 6, 8, 6, 8, 9, 5, 5, 4, 6],
        'Data Engineer': [9, 8, 5, 4, 3, 9, 9, 4, 5, 4, 5],
        'Data Analyst': [9, 6, 7, 4, 3, 4, 4, 9, 6, 5, 8],
        'Quant Analyst': [6, 8, 9, 7, 5, 4, 5, 6, 9, 7, 6]
    }
    
    # Number of variables
    N = len(skills)
    angles = [n / float(N) * 2 * pi for n in range(N)]
    angles += angles[:1]
    
    fig, ax = plt.subplots(figsize=(12, 8), subplot_kw={'projection': 'polar'})
    
    # Draw each role
    colors = {'Data Scientist': 'blue', 'ML Engineer': 'green', 
              'Data Engineer': 'orange', 'Data Analyst': 'purple',
              'Quant Analyst': 'red'}
    
    for role, values in roles.items():
        values_plot = values + values[:1]
        ax.plot(angles, values_plot, 'o-', linewidth=2, 
                label=role, color=colors[role])
        ax.fill(angles, values_plot, alpha=0.1, color=colors[role])
    
    # Add skill labels
    ax.set_xticks(angles[:-1])
    ax.set_xticklabels(skills, size=10)
    ax.set_ylim(0, 10)
    ax.set_yticks([2, 4, 6, 8, 10])
    ax.set_yticklabels(['2', '4', '6', '8', '10'], size=8)
    ax.legend(loc='upper right', bbox_to_anchor=(1.3, 1.0))
    
    ax.set_title('Skills Radar Chart – Financial Data Science Roles', 
                 size=14, pad=20)
    
    plt.tight_layout()
    plt.savefig('skills_radar_chart.png', dpi=300, bbox_inches='tight')
    plt.show()

create_skills_radar()
print("Skills radar chart saved as 'skills_radar_chart.png'")

# ----------------------------------------------------------------
# PART C: CAREER PROGRESSION MATRIX
# ----------------------------------------------------------------

print("\n" + "-"*60)
print("PART C: Career Progression Matrix")
print("-"*60)

career_paths = pd.DataFrame({
    'Level': ['Junior', 'Mid-Level', 'Senior', 'Lead', 'Principal', 'Director'],
    'Data Analyst': ['Junior Analyst', 'Analyst', 'Senior Analyst', 'Lead Analyst', 'Principal Analyst', 'Director of Analytics'],
    'Data Scientist': ['Junior DS', 'Data Scientist', 'Senior DS', 'Lead DS', 'Principal DS', 'Director of DS'],
    'ML Engineer': ['Junior MLE', 'ML Engineer', 'Senior MLE', 'Lead MLE', 'Principal MLE', 'Director of MLOps'],
    'Data Engineer': ['Junior DE', 'Data Engineer', 'Senior DE', 'Lead DE', 'Principal DE', 'Director of DE'],
    'Years Experience': ['0-2', '2-5', '5-8', '8-12', '12-15', '15+'],
    'Key Skills Progression': ['Technical proficiency', 'Project ownership', 'Mentorship', 'Technical leadership', 'Strategic thinking', 'Executive leadership']
})

print("Career Progression Matrix:")
print(career_paths.to_string(index=False))

# ----------------------------------------------------------------
# PART D: TEAM SIZE AND BUDGET BENCHMARKS
# -----------------------------------------------------------------

print("\n" + "-"*60)
print("PART D: Team Size and Budget Benchmarks")
print("-"*60)

benchmarks = pd.DataFrame({
    'Bank Size': ['Small (<$10B assets)', 'Mid ($10-100B)', 'Large ($100B-1T)', 'Mega (>$1T)'],
    'Data Science Team Size': ['5-15', '15-50', '50-200', '200-500'],
    'Data Engineering Size': ['3-10', '10-30', '30-100', '100-300'],
    'MLOps Size': ['0-3', '3-10', '10-30', '30-100'],
    'Governance Size': ['1-3', '3-8', '8-20', '20-50'],
    'Annual Budget ($M)': ['1-5', '5-20', '20-100', '100-300'],
    'Data Scientists/FTE': ['1-2%', '2-4%', '4-6%', '6-10%'],
    'Cloud Spend ($M)': ['0.1-0.5', '0.5-5', '5-20', '20-50']
})

print("\nTeam Size and Budget Benchmarks by Bank Size:")
print(benchmarks.to_string(index=False))

# ----------------------------------------------------------------
# PART E: DATA SCIENCE CULTURE ASSESSMENT
# ----------------------------------------------------------------

print("\n" + "-"*60)
print("PART E: Data Science Culture Assessment")
print("-"*60)

culture_factors = {
    'Factor': ['Leadership Commitment', 'Data Literacy', 'Data Accessibility', 
               'Tooling/Infrastructure', 'Talent/Skills', 'Governance/Quality', 
               'Innovation Culture', 'Cross-functional Collaboration'],
    'Our Score (1-5)': [4, 3, 3, 4, 3, 3, 3, 4],
    'Target Score (1-5)': [5, 4, 4, 5, 4, 4, 4, 5],
    'Gap': [1, 1, 1, 1, 1, 1, 1, 1]
}

culture_df = pd.DataFrame(culture_factors)

# Radar chart for culture assessment
from math import pi
fig, ax = plt.subplots(figsize=(10, 8), subplot_kw={'projection': 'polar'})

factors = culture_df['Factor'].tolist()
our_scores = culture_df['Our Score (1-5)'].tolist()
target_scores = culture_df['Target Score (1-5)'].tolist()

N = len(factors)
angles = [n / float(N) * 2 * pi for n in range(N)]
angles += angles[:1]

our_scores += our_scores[:1]
target_scores += target_scores[:1]

ax.plot(angles, our_scores, 'o-', linewidth=2, label='Current State', color='blue')
ax.fill(angles, our_scores, alpha=0.1, color='blue')
ax.plot(angles, target_scores, 'o-', linewidth=2, label='Target State', color='green', linestyle='--')
ax.fill(angles, target_scores, alpha=0.1, color='green')

ax.set_xticks(angles[:-1])
ax.set_xticklabels(factors, size=9)
ax.set_ylim(0, 5)
ax.set_yticks([1, 2, 3, 4, 5])
ax.set_yticklabels(['1', '2', '3', '4', '5'], size=8)
ax.legend(loc='upper right', bbox_to_anchor=(1.2, 1.0))
ax.set_title('Data Science Culture Assessment', size=14, pad=20)

plt.tight_layout()
plt.savefig('culture_assessment.png', dpi=300, bbox_inches='tight')
plt.show()

# ----------------------------------------------------------------
# PART F: HIRING AND RETENTION STRATEGIES
# ----------------------------------------------------------------

print("\n" + "-"*60)
print("PART F: Hiring and Retention Strategies")
print("-"*60)

print("""
HIRING STRATEGIES:

1. University Recruitment:
   - Partner with top data science and computer science programs.
   - Offer internships and co-op placements.
   - Sponsor data science competitions and hackathons.

2. Experienced Hires:
   - Target fintechs, tech companies (FAANG), and consulting firms.
   - Leverage professional networks (LinkedIn, GitHub, Kaggle).
   - Use specialised recruitment firms for data science.

3. Internal Talent Development:
   - Identify high-potential employees in IT, Risk, and Operations.
   - Provide training programs and certifications.
   - Offer rotation programs across data functions.

4. Diversity & Inclusion:
   - Build diverse candidate pipelines (women in tech, under-represented groups).
   - Implement blind recruitment practices.
   - Foster an inclusive team culture.

RETENTION STRATEGIES:

1. Compensation:
   - Benchmark salaries against market (finance, tech).
   - Offer performance bonuses and equity/stock options.
   - Provide benefits (health, wellness, education).

2. Career Development:
   - Clear career progression pathways.
   - Regular performance reviews and feedback.
   - Support for continuous learning (conferences, certifications).

3. Work Environment:
   - Flexible working arrangements (remote/hybrid).
   - Modern tools and infrastructure.
   - Collaborative and supportive culture.

4. Impact and Recognition:
   - Work on high-impact projects.
   - Publicly recognise achievements.
   - Provide opportunities for publication and speaking.

5. Community:
   - Strong team culture and social events.
   - Mentorship and coaching programs.
   - Contribution to open source and knowledge sharing.
""")

# ----------------------------------------------------------------
# PART G: SUMMARY AND RECOMMENDATIONS
# ----------------------------------------------------------------

print("\n" + "="*70)
print("PART G: Summary and Recommendations")
print("="*70)

print("""
Building a Financial Data Science Team – Key Takeaways:

1. Organisation Models: Centralised, Decentralised, or Hybrid – choose based on culture and scale.
2. Key Roles: Data Scientist, Data Engineer, ML Engineer, Data Analyst, Quant Analyst, Product Manager.
3. Skills: Blend of technical (Python, SQL, ML) and domain (finance, risk) expertise.
4. Culture: Leadership commitment, data literacy, accessibility, and innovation.
5. Hiring: University recruitment, experienced hires, internal upskilling, diversity.
6. Retention: Competitive compensation, career development, flexible work, recognition.
7. Benchmarks: Team size and budget scale with organisational size.

Recommendations:
  - Start with a hybrid model for flexibility and consistency.
  - Invest in data literacy programs across the organisation.
  - Build a strong data engineering foundation before scaling ML.
  - Foster a culture of experimentation and learning.
  - Continuously benchmark against industry peers.
  - Develop clear career paths for data professionals.
""")

print("="*70)
print("END OF LESSON 1 – MODULE 8")
print("="*70)