Decision Intelligence Generative AI Workshop

[Work In Progress - Estimated Beta Launch: December 2025]

Contents:
💡 About
👣 Getting Started: Choose Your Path
📖 Printed Book
🧱 Workshop Modules
✅ Requirements for Interactive Decision Notebooks

About

Navigating decisions effectively is important, serving as a key differentiator for successful individuals. Therefore, decision-making ability is regarded by executives as one of the most highly desired business skills! However, effective and quality decision-making is not simple. Perhaps the intersection of Decision Theory and Generative AI can help.

This Decision Intelligence Generative AI Workshop is an interactive course that is paired with an upcoming book called "Decision Intelligence with Generative AI". This course was designed with a focus on three key areas:

• Illustration of important Decision-Making concepts
• Introduction of the Decision Intelligence Framework that applies systematic decision-making
• Application of Decision Intelligence theory with Generative AI
• Interactive Notebooks for AI engineers/software engineers to execute the examples with an orchestration framework (Semantic Kernel & OpenAI)

The unique workshop structure with diverse focus areas, allows this workshop to be consumed in several ways. Depending on your persona, familiarity with Decision Theory or AI acumen, you have several paths to get started.

Getting Started: Choose Your Path

1. 👓 Reader Persona - Read-Only (Non-Technical) - Ideal for non-technical users not familiar with Decision Intelligence nor Generative AI

• Nothing to install! No special requirements to read any of the workshop content. All of the text, links & images are available by just navigating to the Notebooks folder and reading the section modules sequentially. Notebooks folder: https://github.com/bartczernicki/DecisionIntelligence.GenAI.Workshop/tree/main/Notebooks
• All Generative AI output is pre-run and available to read in a browser
• Recommend further detailed book resources provided in the workshop to immerse yourself in Decision Theory, Decision-Making, Cognitive Theory etc.
• Recommend companion book "Generative AI with Decision Intelligence" (coming late 2025) for a more complete reading experience and advanced content not in the workshop

2. 💻 Technical Persona - Interactive (Technical) - Ideal for technical users that might not be familiar with Decision Intelligence nor Generative AI, but are familiar with Jupyter Notebooks interactivity. The persona can change basic connection strings and make simple code changes.

• All the considerations from the "Read Persona" Getting Started applies
• Set up the Requirements for Interactive Decision Notebooks to add custom interactivity
• Run the workshop Notebook modules sequentially (no coding required!), as the Decision Intelligence & AI concepts "build on" each other
• C#/.NET Code will be executed but simple modifications to decision prompts, agents, configuration files are clear to added decision exploration

3. 🧑‍💻 AI Engineer - Interactive (Advanced Technical) - Ideal for a technical user that is familiar with programming and Generative AI

• Most of the considerations from the "Technical Persona" Getting Started applies
• Set up the Requirements for Interactive Decision Notebooks to add custom interactivity
• Notebook modules do not "build on each other" from a programming dependency; only from a Decision Intelligence concept dependency. Therefore, you can jump in and modify where needed
• Recommended to "Fork and Hack" and gravitate to the Decision Scenarios with most impact (change AI models, try different APIs or knowledge stores, advanced AI programming techniques, port code to your projects etc.). Advanced users are encouraged to fork this repository and try out different models or techniques.

Printed Book

What is the difference between the workshop the printed companion book? The Printed Book has the following features:

• Deeper focus on Decision Intelligence theory with more advanced application of Generative AI rather than a code-first focus.
• Additional advanced real-world examples with Artificial Intelligence.
• Chapters are organized using the Decision Intelligence Framework. A great deal less code illustrated in the book, making it easier to consume for non-technical readers.
• References in text content to further deepen decision-making knowledge.
• Extra Curated Chapters with "Table of Contents Lists" for Easy Lookup of: Decision Intelligence Quotes, Decision Rules, Decision Intelligence Resources.

Workshop Modules

The Table of Contents below illustrates the structure of the Decision Intelligence with Generative AI Workshop. The workshop is structured on concepts broken down into Notebook modules (chapters). Each Notebook module consists of Decision Intelligence concepts and/or interactive Generative AI features that can be dynamically executed.

⚠️ This workshop is currently being developed in conjunction with the companion book. During this time, there will be frequent changes. Content that is available (even if partially) is noted below. Content that does not have the Availability checkmark is in early stages or not available yet. Please check back for updates!

Module (Chapter)	Decision Intelligence	Available	Link
1a - Decision Intelligence - Introducing the Decision Intelligence Framework	Framework Introduction Enhanced with Generative AI Scenario: Application with Eisenhower Decision Priorotization	✅	Link
1b - Decision Intelligence - Decision Framing	Decision Framing Introduction Reframing Alternative Decision Options Systematic Frameworks	✅	Link
1c - Decision Intelligence - Gathering Intelligence	Gathering Intelligence Introduction Historical Example: Battle of Edington Intelligence with AI	✅	Link
1d - Decision Intelligence - Decision Execution	Decision Execution Introduction Three Forms of Execution Can AI Replace a CEO?	✅	Link
1e - Decision Intelligence - Decision Execution with Intuition	Intuition Execution Introduction Scenario: Split-Second Decision Making Intuitive Generative AI	✅	Link
1f - Decision Intelligence - Decision Execution with Decision Rules	Rules Execution Introduction Sample list of Generic Decision Rules Scenario: Exploration vs Exploitation Scenario: Optimize Sales Performance with Price's Law Sample list of Domain Specific (Industry) Decision Rules Customized Rules Frameworks Rules Powered by Generative AI	✅	Link
1g - Decision Intelligence - Decision Execution with Quantitative Methods	Quantitative Execution Introduction Math is Hard for Humans Data Analysis Impossible for Humans Arriving at Quantitative Conclusions using Simulations Introducing the Monte Carlo Scenario: Monte Carlo for Total Cost of Car Ownership	✅	Link
1h - Decision Intelligence - Decision Communication		🛠️	Link
1i - Decision Intelligence - Using the Decision Intelligence Framework		🛠️	Link
1j - Decision Intelligence - Enterprise Decision Intelligence		🛠️	Link
2a - Workshop - Setup to Execute Code Exercises		✅	Link
2b - Workshop - About Microsoft AI Extensions		✅	Link
2c - Workshop - About Semantic Kernel Orchestration		✅	Link
2d - Semantic Kernel - Simple Decision Prompts	Understanding Decision-Making Frameworks Built-Into AI Create Custom Decision-Making AI Personas Scenario: Deciding between Comunity College or a University	✅	Link
2e - Semantic Kernel - Chat Completion for Decisions	Understanding Decision-Making Frameworks Built-Into AI Selecting the Optimal Decision Framework Scenario: Multi-Turn Decision Conversation for a Fitness Device Inspecting Decision Memory with AI	✅	Link
2f - Semantic Kernel - Open Source Decision Intelligence	Scenario: Decision Evaluation for Purchasing an Investment Property Decision Reasoning Decision Reasoning with AI Scenario: Applying the Ben Frankling Decision Framework for Family Vacation Decision	✅	Link
2g - Semantic Kernel - Prompt Engineering for Quality Decisions	Chain of Thought Decision Reasoning Evaluating Chain of Thought Decision Processses Examples of "Wisdom of the Crowds" Collective Intelligence Optimization for Decision-Making Scenario: Applying Collective Intelligence with AI Vision	✅	Link
2h - Semantic Kernel - Premortem (What-If) Decision Analysis	Scenario: Premortem Decision Analysis for a Car Purchase Risk Analysis with AI Reasoning Models Scenario: Premortem Analysis for Radio Telescope Operations	✅	Link
3a - Semantic Kernel - Decisions with Semantic Functions	Dynamic Decisions with Parameters Scenario: Decision Recommendation to Open a Restaurant	✅	Link
3b - Semantic Kernel - Decisions with Native Functions	Simulating Uncertainty using Quantitative Methods	🛠️	Link
3c - Semantic Kernel - Optimizing with Decision Frames	Framing Decisions using Six Thinking Hats Framework Selecting the Optimal Decision Frame	🛠️	Link
4a - Semantic Kernel - Scale Decision Processes with Plugins	Gathering Intelligence for AI Decisions with External Data Stores	🛠️	Link
4b - Semantic Kernel - Plugins for Decision Communication	Decision Communication using the Minto Pyramid Framework Minimizing Judgement During Communication	🛠️	Link
4c - Semantic Kernel - Custom Plugins for Decision Recommendation	Scenario: Decision Support using Internet Gathered Intelligence	🛠️	Link
4d - Semantic Kernel - Diverse Plugins for Decision Making	Decision Execution using various Methods Combining Decision Recommendation using Analytics and Machine Learning Scenario: Predicting Baseball Hall of Fame Induction using Probabilities	🛠️	Link
5a - OpenAI - Improving Decisions with OpenAI LogProbs		🛠️	Link
5b - OpenAI - Measuring GenAI Probabilistic Accuracy with LogProbs		🛠️	Link
5c - OpenAI - Validating Multiple Decisions with Aggregated Brier Scores		🛠️	Link
6a - Semantic Kernel - Decisions with AI Agent Personas	Creating AI Decision Agent Personas Creating AI Dwight Eisenhower Optimizing AI Decision Agents	✅	Link
6b - Semantic Kernel - Decisions with Multi-AI Agent Personas	Decisions with Multi-Agent Personas Decisions with Advanced Multi-Agent Decision Orchestration Scenario: Multi-Agent Investment Decision	✅	Link

Requirements for Interactive Decision Notebooks

1. Visual Studio Code running on your local workstation or VS Code with GitHub CodeSpaces or Azure Machine Learning Notebooks
   • Install VS Code: Download Visual Studio Code
   • Use GitHub CodeSpaces: Open in GitHub CodeSpaces
   • Azure Machine Learning Notebooks
2. Install .NET 9.x SDK:
   • Install .NET 9 SDK: Download .NET 9
3. Polyglot Notebook extension running in VS Code, which allows you to execute .NET program

4. Clone or fork this GitHub Repository 5. Have access to Azure OpenAI. Bing Search for internet knowledge graph (Endpoint, GPT-4 model, API Key, Bing Search API Key) 6. When running the notebooks, select `.NET Interactive` for your VS Code Notebook kernel