This repository contains comprehensive Python Jupyter notebooks for the MBA 590 course, designed to provide hands-on learning experiences for mastering advanced AI interaction paradigms, prompt engineering techniques, and strategic implementation of agentic systems.
This repository now includes comprehensive infrastructure and support materials:
- 📝 4 Assignment Templates - Complete starter files for all major assignments
- 🗂️ Sample Datasets - Real-world data for hands-on practice
- 🎯 Prompt Templates Library - Reusable templates organized by use case
- 🛠️ Python Utilities - Helper functions for LLMs, data analysis, and evaluation
- 🐳 Docker Support - Reproducible environment with one command
- 🧪 Automated Testing - Verify all notebooks execute correctly
- 📚 Comprehensive Documentation - Setup guides, FAQ, troubleshooting
- 🔄 CI/CD Pipeline - GitHub Actions for quality assurance
- 📄 Export Scripts - Convert notebooks to PDF, HTML, slides
data-advanced-ai/
├── notebooks/ # 15 weekly Jupyter notebooks
├── assignments/ # 4 assignment templates
├── data/
│ ├── samples/ # Sample datasets (CSV, JSON)
│ └── examples/ # Example data and prompts
├── prompt_templates/ # Reusable prompt library
├── utils/ # Python utilities module
│ ├── llm_helpers.py
│ ├── data_helpers.py
│ ├── prompt_helpers.py
│ └── evaluation_helpers.py
├── scripts/ # Automation scripts
│ ├── test_notebooks.py
│ └── export_notebooks.py
├── .github/workflows/ # CI/CD configuration
├── Dockerfile # Docker container definition
├── docker-compose.yml # Docker orchestration
├── requirements.txt # Python dependencies
├── environment.yml # Conda environment
├── .env.example # API key template
└── docs/ # Documentation files
This advanced course equips leaders and strategists with the knowledge and skills to effectively leverage sophisticated language models and emerging agentic systems. The curriculum provides an in-depth exploration of:
- Prompt Engineering: From fundamentals to advanced techniques (Chain-of-Thought, RAG, Self-Refine)
- LLM Evaluation: Rigorous frameworks for assessing quality, safety, and business suitability
- Agentic Systems: Architecture, capabilities, and strategic applications
- Implementation Strategy: Tech-ready operating models, governance, and ethical frameworks
- Strategic Leadership: Technology strategy formulation, ROI measurement, and future trends
The course consists of 15 weeks, each with a dedicated Jupyter notebook containing:
- Learning objectives
- Academic readings references
- Conceptual explanations
- Practical Python code examples
- Hands-on exercises
- Discussion questions
- Key takeaways
File: week01_foundations_prompt_engineering.ipynb
Topics:
- Core principles of prompt engineering
- LLM input/output processing
- Context, specificity, and clarity
- Basic prompt types: zero-shot, role-playing, simple instruction
Key Skills: Understanding fundamental prompting principles and comparing different approach styles
File: week02_advanced_prompting_1.ipynb
Topics:
- Few-shot learning for improved context
- Chain-of-Thought (CoT) prompting for complex reasoning
- Self-Ask for problem decomposition
- Combining techniques for optimal results
Key Skills: Implementing sophisticated prompting methods for complex business tasks
File: week03_advanced_prompting_2.ipynb
Topics:
- Least-to-Most (LtM) prompting
- Plan-and-Solve approaches
- Self-Refine for iterative improvement
- Chain-of-Verification (CoVe) for accuracy
Key Skills: Breaking down complex tasks and enabling LLM self-correction
File: week04_rag_prompt_chaining.ipynb
Topics:
- Retrieval-Augmented Generation (RAG) concepts
- Grounding LLMs in specific knowledge
- Prompt chaining for multi-step workflows
- Building complex LLM applications
Key Skills: Implementing RAG and orchestrating multi-step LLM workflows
File: week05_evaluating_llm_outputs.ipynb
Topics:
- Evaluation metrics (BLEU, ROUGE, perplexity, F1-score)
- Quality assessment frameworks
- Bias detection and safety evaluation
- Business suitability criteria
Key Skills: Critically evaluating LLM outputs using established metrics and frameworks
File: week06_intro_agentic_frameworks.ipynb
Topics:
- Defining agentic systems: autonomy, planning, reasoning
- Core concepts: perception, action loops, memory
- ReAct framework and basic architectures
- Tool use and decision-making
Key Skills: Understanding agentic system architecture and capabilities
File: week07_multi_agent_systems.ipynb
Topics:
- Multi-Agent Systems (MAS) concepts
- Communication protocols and patterns
- Coordination and collaboration strategies
- Applications in complex problem-solving
Key Skills: Designing multi-agent systems for business applications
File: week08_agentic_business_applications.ipynb
Topics:
- Real-world applications (customer service, data analysis, supply chain)
- Case studies and implementation examples
- Challenges and success factors
- Strategic opportunities assessment
Key Skills: Evaluating business applications and implementation challenges
Assignment 1 Due: Advanced Prompt Engineering Analysis
File: week09_tech_ready_operating_models_1.ipynb
Topics:
- Organizational structures (centralized, decentralized, CoE)
- Technology governance frameworks
- Roles and responsibilities
- Decision rights and accountability
Key Skills: Designing organizational structures for advanced technology adoption
File: week10_tech_ready_operating_models_2.ipynb
Topics:
- Talent needs and acquisition strategies
- Training and development programs
- Building a culture of experimentation
- Change management and adoption strategies
Key Skills: Developing talent strategies and fostering adaptive cultures
Assignment 2 Due: Agentic Framework Application Proposal
File: week11_governance_ethics_1.ipynb
Topics:
- Ethical risks (bias, fairness, transparency, accountability)
- Responsible AI frameworks (NIST, OECD, IEEE)
- Core principles for ethical deployment
- Policy translation
Key Skills: Applying ethical frameworks to technology governance
File: week12_governance_ethics_2.ipynb
Topics:
- Regulatory landscape (EU AI Act, GDPR)
- Bias audits and testing
- Transparency reporting and model cards
- Human oversight mechanisms
Key Skills: Implementing governance controls and regulatory compliance
Assignment 3 Due: Tech-Ready Operating Model Design
File: week13_technology_strategy.ipynb
Topics:
- Technology strategy frameworks
- Project prioritization methodologies
- Portfolio management and balance
- Roadmap development
Key Skills: Formulating comprehensive technology strategies
File: week14_measuring_roi.ipynb
Topics:
- ROI calculation fundamentals (NPV, IRR, Payback Period)
- Total Cost of Ownership (TCO) analysis
- Tangible and intangible benefits
- Balanced Scorecard for technology
Key Skills: Calculating and assessing technology initiative ROI
File: week15_future_trends.ipynb
Topics:
- Emerging technology trends (multimodal AI, AGI, quantum)
- Future scenarios and strategic planning
- Leadership capabilities for the AI era
- Course synthesis and personal action planning
Key Skills: Assessing future trends and developing strategic leadership capabilities
Assignment 4 Due: AI Strategic Implementation Plan
Access pre-built prompts for common business tasks:
from prompt_templates.business_templates import swot_analysis
# Fill in a template
prompt = swot_analysis.fill(
context="Entering the AI automation market",
subject="Our AI Platform",
market="Mid-market B2B SaaS"
)Available Templates:
- Business Analysis (SWOT, competitive analysis, market opportunities)
- Customer Service (complaint handling, product inquiries)
- Content Creation (product descriptions, marketing copy)
- Data Analysis (trend analysis, anomaly detection)
- Technical (code review, API documentation)
See prompt_templates/README.md for full documentation.
Helper functions for common tasks:
from utils import count_tokens, estimate_cost, create_mock_llm_response
from utils import load_sample_data, create_comparison_chart
from utils import validate_prompt, measure_prompt_quality
# Count tokens and estimate costs
tokens = count_tokens("Your prompt here")
cost = estimate_cost(tokens, expected_output=500, model="gpt-4")
# Load sample data
df = load_sample_data("customer_service_tickets.csv")
# Validate prompts
quality = validate_prompt("Your prompt")
print(f"Quality score: {quality['score']}/100")See inline documentation for all available functions.
Four comprehensive assignment templates aligned with course milestones:
- Assignment 1 (Week 8): Advanced Prompt Engineering Analysis - 20%
- Assignment 2 (Week 10): Agentic Framework Application Proposal - 20%
- Assignment 3 (Week 12): Tech-Ready Operating Model Design - 20%
- Assignment 4 (Week 15): AI Strategic Implementation Plan - 30%
Each includes:
- Detailed instructions
- Code scaffolding
- Evaluation rubrics
- Submission checklists
We support three deployment options to fit your needs:
| Option | Best For | Setup Time | Cost | Privacy |
|---|---|---|---|---|
| 🏢 Azure Cloud | Enterprise, compliance needs | 15-30 min | $$ | High (VNet) |
| 🖥️ Local LLM | Privacy, offline, no API costs | 10-20 min | Free | Complete |
| ☁️ OpenAI Direct | Quick start, best quality | 5 min | $ | Standard |
Enterprise-grade deployment with security, compliance, and VNet integration.
# See complete guide
cat DEPLOY_AZURE.md
# Quick deploy with Azure CLI
az group create --name mba590-rg --location eastus
az deployment group create \
--resource-group mba590-rg \
--template-file azure/deploy-azure-openai.bicep
# Copy credentials to .env
cp .env.example .env
# Set LLM_PROVIDER=azure and add your Azure credentialsBenefits:
- ✅ SOC 2, HIPAA, ISO compliance
- ✅ Private network (VNet) integration
- ✅ Data residency control
- ✅ Enterprise SLA (99.9%)
- ✅ Cost tracking and quotas
Run completely locally with no cloud API required.
# See complete guide
cat DEPLOY_LOCAL.md
# Quick start with Docker
docker-compose -f docker-compose.local-llm.yml up
# Or install Ollama directly
curl -fsSL https://ollama.ai/install.sh | sh
ollama pull llama2:7b
# Start Jupyter
cp .env.example .env
# Set LLM_PROVIDER=ollama
docker-compose upBenefits:
- ✅ Complete data privacy (nothing leaves your machine)
- ✅ No API costs (free to use)
- ✅ Works offline
- ✅ Full model control
- ✅ Open-source models
Fastest way to get started:
# Clone the repository
git clone https://github.com/yourusername/data-advanced-ai.git
cd data-advanced-ai
# Copy environment template
cp .env.example .env
# Add your OpenAI API key to .env
# LLM_PROVIDER=openai
# OPENAI_API_KEY=your_key_here
# Start with Docker
docker-compose up
# Access Jupyter at http://localhost:8888- Python 3.8 or higher
- Jupyter Notebook or JupyterLab
- Basic understanding of Python programming
- Familiarity with business strategy concepts
-
Clone the repository:
git clone https://github.com/yourusername/data-advanced-ai.git cd data-advanced-ai -
Create a virtual environment (recommended):
python -m venv venv source venv/bin/activate # On Windows: venv\Scripts\activate
-
Install required packages:
pip install jupyter pandas numpy matplotlib seaborn plotly scikit-learn
-
Optional - Install LLM API libraries (for running live examples):
pip install openai anthropic
-
Launch Jupyter:
jupyter notebook
or
jupyter lab
-
Navigate to the notebooks folder and open the desired week's notebook
-
Follow the instructions in each notebook, running cells sequentially
-
Advanced Prompt Engineering Analysis (20%) - Due Week 8
- Apply advanced prompting techniques to a real business scenario
- Evaluate effectiveness using multiple approaches
-
Agentic Framework Application Proposal (20%) - Due Week 10
- Design an agentic system for a specific business use case
- Include architecture, capabilities, and implementation plan
-
Tech-Ready Operating Model Design (20%) - Due Week 12
- Develop organizational structure and governance for AI adoption
- Address talent, culture, and change management
-
AI Strategic Implementation Plan (30%) - Due Week 15
- Comprehensive technology strategy document
- Include prioritization, roadmap, ROI analysis, and governance
- Active participation in discussion forums
- Completion of hands-on exercises in notebooks
Upon successful completion of this course, you will be able to:
- Master Advanced Prompt Engineering: Apply sophisticated prompting methods for complex tasks
- Critically Evaluate LLM Outputs: Use metrics and frameworks to assess quality and suitability
- Analyze Agentic Systems: Understand architectures and capabilities
- Evaluate Strategic Applications: Identify opportunities and assess risks
- Design Tech-Ready Operating Models: Create organizational structures and cultural attributes
- Implement Governance: Develop ethical and responsible technology frameworks
- Formulate Technology Strategies: Create comprehensive strategies aligned with business goals
- Measure ROI: Apply methodologies to assess technology initiative value
- Assess Future Trends: Analyze emerging technologies and strategic implications
- Sequential Learning: Work through notebooks in order, as concepts build on previous weeks
- Hands-On Practice: Complete all "YOUR TURN" exercises to reinforce learning
- Experimentation: Modify code examples to explore different scenarios
- Discussion Participation: Engage with discussion questions and share insights
- Real-World Application: Apply concepts to your own business context
- Resource Exploration: Review additional resources provided in each notebook
- Most code examples are self-contained and can run without external APIs
- Some examples demonstrate API integration (OpenAI, Anthropic) but include placeholder code
- Visualization libraries used: matplotlib, seaborn, plotly
- Data analysis with: pandas, numpy, scikit-learn
If you want to run live LLM examples:
-
Obtain API keys from:
- OpenAI: https://platform.openai.com/
- Anthropic: https://console.anthropic.com/
-
Set environment variables:
export OPENAI_API_KEY='your-key-here' export ANTHROPIC_API_KEY='your-key-here'
-
Or create a
.envfile in the project root (included in .gitignore)
Each notebook references academic papers and industry reports. Key papers include:
- Wei et al. (2022): Chain-of-Thought Prompting
- Lewis et al. (2020): Retrieval-Augmented Generation
- Xi et al. (2023): Large Language Model Based Agents Survey
- Madaan et al. (2023): Self-Refine
- Zhou et al. (2022): Least-to-Most Prompting
- Chang et al. (2023): Evaluation of Large Language Models
- NIST (2023): AI Risk Management Framework
Full references are provided within each notebook.
Verify all notebooks execute correctly:
# Test all notebooks
python scripts/test_notebooks.py
# Test specific directory
python scripts/test_notebooks.py --dir notebooks
# With custom timeout
python scripts/test_notebooks.py --timeout 300Convert notebooks to various formats:
# Export to HTML
python scripts/export_notebooks.py --formats html
# Export to multiple formats
python scripts/export_notebooks.py --formats html pdf slides
# Custom output directory
python scripts/export_notebooks.py --output my_exports# Format code
black utils/ prompt_templates/
# Lint code
pylint utils/ prompt_templates/
# Type checking
mypy utils/- README.md - This file, repository overview
- CONTRIBUTING.md - How to contribute
- FAQ.md - Frequently asked questions
- TROUBLESHOOTING.md - Common issues and solutions
- CHANGELOG.md - Version history and updates
- prompt_templates/README.md - Prompt library guide
We welcome contributions from students, instructors, and the community!
Ways to Contribute:
- Report bugs or errors
- Suggest improvements
- Add new examples
- Update documentation
- Share interesting use cases
See CONTRIBUTING.md for detailed guidelines.
Quick contribution workflow:
- Fork the repository
- Create a feature branch
- Make your changes
- Test thoroughly
- Submit a pull request
This course material is provided for educational purposes. Please refer to your institution's academic policies regarding use and distribution of course materials.
Course: MBA 590 - Advanced AI Strategy: Prompting and Agentic Frameworks Institution: NC State University, Poole College of Management Term: Spring 2026 Credit Hours: 3 Format: Online with synchronous office hours
Instructor Information: See course syllabus on WolfWare
- Technical Issues: Contact OIT Help Desk at 919-515-HELP or help@ncsu.edu
- Course Content Questions: Use the discussion forum or attend office hours
- Notebook Issues: Open an issue in this repository
Official course materials and submissions: https://wolfware.ncsu.edu/courses/details/?sis_id=SIS:2026:1:1:MBA:590:639
| Week | Topic | Notebook File | Assignment Due |
|---|---|---|---|
| 1 | Prompt Engineering Foundations | week01_foundations_prompt_engineering.ipynb | - |
| 2 | Advanced Prompting I | week02_advanced_prompting_1.ipynb | - |
| 3 | Advanced Prompting II | week03_advanced_prompting_2.ipynb | - |
| 4 | RAG & Prompt Chaining | week04_rag_prompt_chaining.ipynb | - |
| 5 | Evaluating LLM Outputs | week05_evaluating_llm_outputs.ipynb | - |
| 6 | Intro to Agentic Frameworks | week06_intro_agentic_frameworks.ipynb | - |
| 7 | Multi-Agent Systems | week07_multi_agent_systems.ipynb | - |
| 8 | Agentic Business Applications | week08_agentic_business_applications.ipynb | Assignment 1 |
| 9 | Operating Models I | week09_tech_ready_operating_models_1.ipynb | - |
| 10 | Operating Models II | week10_tech_ready_operating_models_2.ipynb | Assignment 2 |
| 11 | Governance & Ethics I | week11_governance_ethics_1.ipynb | - |
| 12 | Governance & Ethics II | week12_governance_ethics_2.ipynb | Assignment 3 |
| 13 | Technology Strategy | week13_technology_strategy.ipynb | - |
| 14 | Measuring ROI | week14_measuring_roi.ipynb | - |
| 15 | Future Trends & Leadership | week15_future_trends.ipynb | Assignment 4 |
Last Updated: November 2025 Version: 1.0
Happy Learning! 🚀