AI Project Dashboard

An enterprise-grade AI project portfolio management dashboard that empowers organizations to track, analyze, and optimize all their AI initiatives from a single unified interface. Built with Streamlit, Pydantic, and SQLite for rapid deployment with zero infrastructure overhead.

Key Features

• Portfolio Overview -- Visualize all AI projects with status indicators, priority levels, and health scoring
• KPI Tracking -- Monitor key performance indicators with trend analysis and achievement rate calculations
• Budget Management -- Track planned vs actual spending with variance analysis by project and category
• Risk Register -- Interactive risk matrix heatmap with probability/impact scoring and mitigation tracking
• Executive Summary -- Auto-generated executive reports with actionable recommendations
• Portfolio Health Score -- Composite 0-100 health metric across status, risk, budget, and KPI dimensions
• ROI Calculator -- Compute return on investment from budget actuals and KPI achievement data
• Report Generation -- Export Markdown/HTML reports for portfolio overview, budget variance, and risk register

Architecture

+-------------------+     +-------------------+     +-------------------+
|   Streamlit UI    |     |   Report Engine   |     |   Analytics       |
|   (app.py)        |---->|   (report_gen.py) |     |   (analytics.py)  |
|                   |     +-------------------+     |                   |
|  - Portfolio View |            |                  |  - ROI Calculator |
|  - KPI Tracking   |            |                  |  - Health Score   |
|  - Budget Mgmt    |            v                  |  - Trend Analyzer |
|  - Risk Register  |     +-------------------+     |  - Exec Summary   |
|  - Exec Summary   |---->|   Database Layer  |<----|                   |
+-------------------+     |   (database.py)   |     +-------------------+
                          |                   |
                          |  SQLite Backend   |
                          +-------------------+
                                  |
                          +-------------------+
                          |  Pydantic Models  |
                          |  (models.py)      |
                          |                   |
                          |  - AIProject      |
                          |  - ProjectKPI     |
                          |  - BudgetEntry    |
                          |  - RiskEntry      |
                          +-------------------+

Quick Start

Prerequisites

• Python 3.12 or higher
• uv package manager (recommended)

Installation

# Clone the repository
git clone https://github.com/onurcandonmezer/ai-project-dashboard.git
cd ai-project-dashboard

# Create virtual environment and install
uv venv
uv pip install -e ".[dev]"

Run the Dashboard

# Run with Streamlit
uv run streamlit run src/app.py

# Or use the Makefile
make run

The dashboard will automatically seed sample data on first run.

Run Tests

# Run all tests
uv run python -m pytest tests/ -v --tb=short

# Run with coverage
make coverage

Usage Examples

Programmatic Access

from src.database import ProjectDatabase
from src.models import AIProject, ProjectStatus, Priority
from src.analytics import PortfolioHealthScore, ROICalculator
from datetime import date

# Initialize database
db = ProjectDatabase("my_portfolio.db")

# Add a project
project = AIProject(
    name="Customer Churn Predictor",
    description="ML model to predict customer churn",
    status=ProjectStatus.DEVELOPMENT,
    priority=Priority.HIGH,
    owner="Data Science Team",
    start_date=date(2024, 6, 1),
    target_date=date(2025, 1, 31),
    model_used="XGBoost",
    use_case="Churn Prediction",
    department="Marketing",
)
db.add_project(project)

# Compute portfolio health
projects = db.get_all_projects()
health = PortfolioHealthScore.compute(
    projects,
    db.get_all_risks(),
    db.get_all_budgets(),
    db.get_all_kpis(),
)
print(f"Portfolio Health: {health.overall_score}/100")

Generate Reports

from src.report_generator import ReportGenerator

report = ReportGenerator.executive_summary_report(
    projects, kpis, budgets, risks
)
print(report)

# Export as HTML
html = ReportGenerator.to_html(report, title="Q4 AI Portfolio Review")

Seed from YAML

make seed

Tech Stack

Component	Technology
Frontend	Streamlit, Plotly
Data Models	Pydantic v2
Database	SQLite (WAL mode)
Data Processing	Pandas, NumPy
Configuration	PyYAML
CLI Output	Rich
Testing	pytest, pytest-cov
Linting	Ruff
Package Management	uv

Project Structure

ai-project-dashboard/
├── README.md
├── pyproject.toml
├── Makefile
├── LICENSE
├── .gitignore
├── .github/
│   └── workflows/
│       └── ci.yml
├── src/
│   ├── __init__.py
│   ├── models.py            # Pydantic data models
│   ├── database.py          # SQLite database layer
│   ├── analytics.py         # ROI, health score, trends
│   ├── report_generator.py  # Markdown/HTML reports
│   └── app.py               # Streamlit dashboard
├── tests/
│   ├── __init__.py
│   ├── test_models.py       # Model validation tests
│   ├── test_database.py     # CRUD and query tests
│   └── test_analytics.py    # Analytics logic tests
├── data/
│   └── sample_projects.yaml # Sample AI project data
└── assets/

Sample Data

The project includes 8 realistic AI projects across departments:

Project	Status	Department
Enterprise AI Chatbot	Production	Customer Success
Product Recommendation Engine	Production	E-Commerce
Real-time Fraud Detection	Testing	Finance & Risk
Intelligent Document Processing	Development	Operations
Demand Forecasting Platform	Planning	Supply Chain
Brand Sentiment Analyzer	Production	Marketing
AI Code Review Assistant	Development	Engineering
Legacy NLP Classifier	Retired	Customer Success

License

This project is licensed under the MIT License - see the LICENSE file for details.

Copyright (c) 2024 Onurcan Donmezer