Programstrategyhq.

com 1
 Foreword

In the ever-evolving landscape of technology and management, the emergence of Generative AI (GenAI)
stands as a transformative milestone, heralding a new era of possibilities and challenges. "Generative AI
for Non-Technical Managers" is a handy book designed with a clear and vital purpose: to demystify the
complexities of GenAI for those who may not have a technical background but are keen to harness its
potential in their managerial roles.

This short book is intentionally crafted to be simple, accessible, and engaging. It aims to bridge the gap
between high-level technical jargon and the practical understanding needed by managers and leaders in
various industries. We believe that the true power of GenAI lies not just in the hands of technologists and
data scientists, but equally in the strategic vision of managers who can integrate these advancements into
their workflows, teams, and projects.

As you delve into the pages of this book, you will find concepts broken down into easily digestible segments,
accompanied by real-world examples, and a focus on practical application. Whether you're leading a team
in a tech company, managing projects building GenAI capabilities or any other sector, this book offers you
the insights to understand and leverage GenAI in your domain.

We hope that by the end of this journey, you will not only have a foundational understanding of Generative
AI but also be equipped with the knowledge to make informed decisions, drive innovation, and lead your
teams with confidence in an AI-augmented future.

Welcome to the world of Generative AI. Let's embark on this enlightening journey together.

Programstrategyhq.com 2
 Generative AI for Project Managers

Table of Contents
Part 1: What is Generative AI?.......................................................................................................... 4
AI ML Basics ...................................................................................................................................... 6
What are LLMs?.............................................................................................................................. 10
How does a GenAI application work? ...................................................................................... 14
Part 2: Technology for GenAI Projects ......................................................................................... 17
Create a GenAI Assist Chatbot ................................................................................................... 17
Other Technical Considerations................................................................................................. 20
GenAI Common Terms.................................................................................................................. 22
Part 3: GenAI for Project Managers .............................................................................................. 26
GenAI Tools ..................................................................................................................................... 26
Prompt Engineering ...................................................................................................................... 30
GenAI Use cases in Project Management ............................................................................... 35
How to Plan Gen AI Projects ....................................................................................................... 38
Gen AI Project Metrics .................................................................................................................. 39
Resources – Gen AI Online Courses & Books ....................................................................... 41
References ........................................................................................................................................... 45

Programstrategyhq.com 3
Part 1: What is Generative AI?
Generative artificial intelligence (GenAI) is artificial intelligence capable of generating text,
images, or other data using generative models, often in response to prompts. Generative AI
models learn the patterns and structure of their input training data and then generate new data
that has similar characteristics.

The concepts of AI, ML, Deep Learning, LLMs, and Foundation Models collectively form the
bedrock of Generative AI (GenAI). While AI provides the overarching framework for intelligent
systems, ML and Deep Learning contribute the methodologies and structures, such as neural
networks, for these systems to learn and adapt. LLMs and Foundation Models, especially those
utilizing transformers and retrieval-augmented generation, advance this further by enabling the
processing and generation of human-like text.

GenAI stands at the convergence of these technologies, harnessing their combined strengths to
create intelligent applications that can generate novel, relevant, and contextually rich content.
This technology in project management can streamline processes, foster innovation, and drive
efficient decision-making among many other applications. Before getting inro more details on
GenAI, lets begin with the fundamentals of AI/ML.

Programstrategyhq.com 4
Additional Reading: Check out this wonderful 18-minute video by Henrik Kniberg - Generative
AI in a Nutshell - how to survive and thrive in the age of AI.

Programstrategyhq.com 5
AI ML Basics

AI (Artificial Intelligence)
Definition: Artificial Intelligence refers to the simulation of human intelligence in machines
programmed to mimic human thought processes and behavior.

Use Case: Chatbots in customer service utilize AI to interpret and respond to customer inquiries,
improving efficiency and user experience.

ML (Machine Learning)
Definition: Machine Learning is a subset of AI focusing on the development of algorithms that
enable machines to learn and improve from experience.

Use Case: Recommendation systems in online platforms, like Netflix, use ML to analyze viewing
habits and suggest personalized content to users.

Deep Learning
Definition: Deep Learning is a subset of ML based on artificial neural networks with
representation learning. It learns from vast amounts of unstructured data.

Use Case: Voice assistants like Siri or Google Assistant use deep learning to understand and
process natural language queries.

GenAI (Generative AI)

Definition: Generative AI is an advanced area within AI, where algorithms create new and original
outputs, ranging from text to images, based on learned data.

It is an extension of deep learning and ML, leveraging complex models to generate novel outputs
instead of just interpreting or classifying data.

GenAI Types and Use Cases

Generative AI, characterized by its ability to generate new content, encompasses various types
and has a wide range of applications. Understanding these types and their potential use cases is
crucial for program managers overseeing projects in this domain.

Programstrategyhq.com 6
1. Text Generation Models
• Description: These models generate textual content, which can range from simple
responses to complex narrative texts.
• Use Cases:
• Content creation for blogs, articles, and social media posts.
• Automated report generation.
• Scriptwriting and narrative design in gaming.

2. Image Generation Models

• Description: Capable of creating visual content, these models generate new images
or modify existing ones.
• Use Cases:
• Generating artwork and designs.
• Visual data augmentation for training other AI models.
• Concept art creation for media and entertainment industries.

3. Music and Sound Generation Models

• Description: These models create musical or audio content, either by composing new
pieces or altering existing tracks.
• Use Cases:
• Composing background scores for games and films.
• Creating unique sound effects.
• Assisting in music production by generating melodies and harmonies.

4. Data Generation Models

• Description: Focused on generating synthetic data, these models are used to create
datasets that mimic real-world data.
• Use Cases:
• Data augmentation for training machine learning models.
• Generating test data for software testing and validation.
• Research in fields where data privacy is a concern, like healthcare.

5. Video Generation Models

Programstrategyhq.com 7
 • Description: These models can generate or alter video content, creating new footage
or modifying existing videos.
• Use Cases:
• Creating training videos for educational purposes.
• Generating visual effects in the film industry.
• Developing marketing and promotional content.

6. 3D Model Generation
• Description: Capable of generating three-dimensional models, these are used in
various design and simulation applications.
• Use Cases:
• Architectural design and visualization.
• Creation of 3D models for video games and virtual reality.
• Prototyping in product design and manufacturing.

For program managers, understanding these types and their applications is vital. Each type of
Generative AI brings unique project challenges, such as varying computational requirements,
data privacy considerations, ethical implications, and the necessity for specialized talent.

Differences between AI, ML, DL and GenAI

Artificial Machine Learning

Aspect Intelligence (AI) (ML) Deep Learning Generative AI

A broad field of
computer science
aimed at building A subset of AI focused A subset of ML
smart machines on the development of that uses neural A type of AI
capable of algorithms that can networks with focused on
performing tasks learn and make many layers creating new
that typically predictions or (deep content, data, or
require human decisions based on networks) to information based
Definition intelligence. data. learn from data. on training data.

Programstrategyhq.com 8
 Artificial Machine Learning
Aspect Intelligence (AI) (ML) Deep Learning Generative AI

Uses complex Uses advanced

neural networks algorithms, often
for pattern based on deep
Uses statistical recognition and learning, to
Uses algorithms methods to enable learning from generate new data
based on logic and machines to improve large amounts similar to the
Approach rules. with experience. of data. training set.

Image and
speech
Virtual assistants, Recommendation recognition, Art generation, text
game playing, systems, spam natural generation, music
language filtering, fraud language composition, data
Applications translation, etc. detection, etc. processing, etc. augmentation, etc.

Varies widely; Requires a large

rule-based and diverse
systems require Needs large dataset to
less data, volumes of data effectively
learning-based Requires substantial for effective generate high-
Data systems need data for training and learning and quality, novel
Dependency more. improving accuracy. accuracy. outputs.

Very high, as it
needs to
Ranges from High complexity understand and
simple, rule-based Generally more due to deep replicate the
systems to complex than neural networks nuances of the
complex learning traditional algorithms and large-scale input data to
and problem- but simpler than deep data generate new
Complexity solving. learning. processing. content.

Programstrategyhq.com 9
What are LLMs?

What are LLMs?

Definition: Large Language Models (LLMs) are advanced AI models designed to understand,
interpret, generate, and respond to human language on a large scale.

LLMs are a prominent example of Generative AI (GenAI) in the field of natural language
processing (NLP). They utilize the principles of machine learning and deep learning to generate
human-like text, making them a significant subset of AI technologies.

1. Autoregressive Models (e.g., GPT-3):

• Application: Content generation, language translation, creative writing.
• Example: OpenAI's GPT-3.

2. Autoencoding Models (e.g., BERT):

• Application: Text summarization, question answering, information extraction.
• Example: Google's BERT.

3. Seq2Seq Models (e.g., T5):

• Application: Machine translation, summarization, text-to-speech.
• Example: Google’s T5 (Text-to-Text Transfer Transformer).

What are Foundation Models?

Foundation models are versatile, deep learning models pre-trained on extensive, diverse
datasets. They provide a base that can be fine-tuned for various specific tasks, including but not
limited to language, vision, and decision-making processes. Examples include

1. GPT-3 (OpenAI):
• An autoregressive language model known for its ability to generate coherent and
contextually relevant text.
• Applications: Content creation, conversational agents, language translation, and
more.

Programstrategyhq.com 10
2. BERT (Bidirectional Encoder Representations from Transformers - Google):
• A model designed to understand the context of a word in a sentence by examining
the words around it.
• Applications: Text classification, search engine optimization, sentiment analysis,
question answering.

3. T5 (Text-To-Text Transfer Transformer - Google):

• Frames all NLP tasks as text-to-text problems, converting every NLP problem into
a text generation task.
• Applications: Machine translation, summarization, question answering, text
classification.

4. RoBERTa (Robustly Optimized BERT Pretraining Approach - Facebook AI):

• An optimized version of BERT, pre-trained on a larger dataset and for a longer
time.
• Applications: Enhanced versions of BERT’s applications like natural language
inference, sentiment analysis.

5. LaMDA (Language Model for Dialogue Applications - Google):

• Specialized in dialogue, this model is designed to engage in free-flowing
conversations on a seemingly endless number of topics.
• Applications: Conversational AI, chatbots, virtual assistants.

6. LLaMA (Large Language Model - Meta/Facebook AI):

• A model emphasizing accessibility and efficiency, suitable for a range of hardware
environments.
• Applications: Flexible deployment in various environments, including those with
limited computational resources.

7. PaLM (Pathways Language Model - Google):

• A model focused on multi-tasking and understanding a vast range of languages
and tasks.

Programstrategyhq.com 11
 • Applications: Multilingual language understanding, complex task handling, and
multi-modal applications.

How does the LLM process information?

This below sequence illustrates the process through which a language model operates, from
receiving an input to providing an output. This involves machine learning algorithms and neural
network processes, with high computational powers

1. Input Processing
• Step 1: User inputs a query or statement.
• Step 2: The input is preprocessed to convert it into a format understandable by the
model. This typically involves tokenization, where the input text is broken down
into smaller pieces, often called tokens.

2. Model Interaction
• Step 3: The preprocessed input is fed into the language model.
• Step 4: The model, which consists of multiple layers of neural networks, processes
the input. Each layer performs specific transformations and feature extractions.

3. Context Understanding and Response Generation

Programstrategyhq.com 12
 • Step 5: The model analyzes the input considering the context, patterns, and
training it has received.
• Step 6: Based on this analysis, the model generates a response. This involves
predicting the next words or sentences that best respond to the input, based on its
training data.

4. Output Processing
• Step 7: The generated response is post-processed if necessary, to ensure it is in
a human-readable format.
• Step 8: The response is delivered to the user.

5. Feedback Loop (Optional)

• Step 9: In some implementations, feedback from the user on the quality of the
response is used for continuous learning and model improvement.

There is a distinction between a service and a model in the context of cloud computing. A
service, as provided by Cloud Service Providers, represents a comprehensive offering that
typically amalgamates various elements such as models, data, and additional components. On
the other hand, a model acts as the pivotal element within a service, often grounded in a
foundational model like a Large Language Model (LLM).

Services are generally fine-tuned for production environments, with a focus on user-friendliness,
often facilitated through a graphical user interface. Its important to note that these services are
often associated with a cost. They could necessitate a subscription or other forms of payment.
One example of a service is Azure Machine Learning which is a Cloud Service designed for data
scientists and ML engineers to manage the whole ML lifecycle (train, test, deploy and handle
MLOps) in a single platform.

Programstrategyhq.com 13
How does a GenAI application work?

Tech stack - Key Components of Software Architecture

A Generative AI (GenAI) application, such as a chatbot agent, relies on a specific set of
technologies. This tech stack comprises various components that work together to enable the
functionality of the GenAI application. The key components include:

1. User Interface (UI): This is the front-end part where user interaction happens. It's
designed to be user-friendly and intuitive.
• Technologies: HTML, CSS, JavaScript, and front-end frameworks like React or
Angular.

2. Application Programming Interface (API): APIs act as intermediaries, allowing the

user interface to communicate with the server and the GenAI model.
• Technologies: RESTful APIs, GraphQL.

3. GenAI Model: The core of a GenAI application, this is where the generative AI
algorithm resides. Large language models like GPT-3, BERT, or custom-built
models using TensorFlow or PyTorch.

4. Data Storage: This component is responsible for storing data that the GenAI
application uses and generates.
• Technologies: Databases such as PostgreSQL, MongoDB, or cloud storage
solutions like AWS S3, Google Cloud Storage.

5. Server and Runtime Environment: Servers host the back-end logic and the runtime
environment where the GenAI model operates.
• Technologies: Node.js, Python Flask/Django; server infrastructure like AWS EC2,
Google Cloud Compute Engine.

6. Machine Learning Operations (MLOps) Infrastructure: Essential for training,

updating, and maintaining the GenAI models.

Programstrategyhq.com 14
 • Technologies: Tools like Kubernetes, Docker, and CI/CD pipelines for automated
deployment and scaling.

7. Security and Compliance: Ensures the application is secure and complies with data
protection laws.
• Technologies: SSL/TLS for encryption, OAuth for authentication, and compliance
tools for GDPR, HIPAA, etc.

How a GenAI Application Works in Realtime: Non-Technical Explanation

Consider a GenAI application, such as a chatbot agent, as a digital concierge in a hotel. Here's a
simplified explanation of how it operates in real-time:

1. Interaction Initiation:
Just like a guest approaching a concierge, a user starts by inputting a query or request
through the application’s user interface (the digital equivalent of a hotel lobby).

2. Request Processing:
The request is sent to the server via an API, much like a concierge listening to the guest's
request. The server is the back-office where all the processing happens.

3. GenAI Model Engagement:

The core GenAI model (the brain of the concierge) then interprets the request. This model,
trained on vast amounts of data, understands and generates a response, just like a
knowledgeable concierge crafting a reply based on their experience.

4. Response Delivery:
This response is then sent back to the user through the UI, providing information or
assistance, similar to how a concierge would give suggestions or solutions to a guest.

5. Continuous Learning:
Just as a concierge learns from each interaction to improve service, the GenAI model
continuously learns from interactions to enhance its responses over time.

Programstrategyhq.com 15
In this way, the GenAI application provides real-time, intelligent responses to user queries,
leveraging advanced AI and a robust tech stack to deliver an experience akin to interacting with
an expert human agent.

Programstrategyhq.com 16
Part 2: Technology for GenAI Projects

Create a GenAI Assist Chatbot

The integration of conversational AI chatbots on websites is rapidly becoming a norm. Lets see
how to create a Conversational GenAI Chatbot for the website programstrategyhq.com for
Enhanced User Engagement.

ProgramStrategyHQ.com website, aiming to stay at the forefront of this technological wave, plans
to develop an automated conversational bot. This bot will not only help in topic discovery and AI
search but will also enhance website navigation, providing a seamless user experience. This blog
post outlines the strategic approach for creating such a bot, tailored to meet the needs of our
audience.

1. Defining Components and Workflow Design

To embark on creating an AI chatbot, it's crucial to understand its core components:

• User Interface (UI): The front-end through which users will interact with the bot.
• Natural Language Processing (NLP) Engine: The backend technology that interprets
user input and generates responses.
• Data Storage: For storing conversation logs and user data, crucial for continuous learning
and improvement.

The workflow for creating the bot involves:

1. Requirement Analysis: Understanding the specific needs of ProgramStrategyHQ.com

and its users.
2. Designing the Conversation Flow: Mapping out potential conversations to ensure a
natural and helpful user experience.
3. Selecting the Right Technology Stack: Based on the analysis and design, choosing the
appropriate tools and platforms.

2. Exploring and Choosing Large Language Models (LLMs)

There are several LLMs available, like OpenAI's GPT-4, Google's BERT, and others. The
selection depends on factors such as:

Programstrategyhq.com 17
 • Model Performance: Accuracy in understanding and responding to queries.
• Scalability: Ability to handle multiple queries simultaneously.
• Cost-Effectiveness: Balancing budget and performance needs.

3. Applying Prompting Techniques

To ensure effective communication, the chatbot must be skilled in:

• Asking Relevant Questions: Crafting prompts that guide users to articulate their needs
accurately.
• Evaluating Responses: Interpreting user input correctly to provide pertinent information
or actions.

4. Establishing Performance Metrics

Key metrics to measure model performance include:

• Accuracy: The rate at which the bot provides correct and relevant responses.
• User Satisfaction: Through feedback forms and direct user ratings.
• Response Time: Ensuring prompt replies for enhanced user experience.

5. Prompt Engineering
Improving the assistant's responses involves:

• Simple Prompting Techniques: Using straightforward prompts to direct the

conversation.
• Chain of Thought Reasoning-Based Prompting: This involves the bot breaking down
complex queries into simpler components for better understanding and response
accuracy.

6. Fine-Tuning Using OpenAI APIs

Customization is key. Fine-tuning the chosen LLM using OpenAI APIs involves:

• Training on Custom Data: Incorporating specific data from ProgramStrategyHQ.com to

make the bot more aligned with our content and audience.
• Iterative Testing and Improvement: Continuously refining the model based on test
results and user feedback.

Programstrategyhq.com 18
7. Deployment and Launch
The final phase includes:

• Integrating the Bot with the Website: Ensuring seamless UI integration.

• Beta Testing: A pilot launch with a select user group to gather initial feedback.
• Full Launch: Post successful beta testing, rolling out the bot to all users with continuous
monitoring and improvements.

Creating a conversational AI chatbot for ProgramStrategyHQ.com involves a structured

approach, combining the right technology with user-centric design. This initiative not only propels
us into the realm of advanced digital interaction but also significantly enhances user engagement
and satisfaction.

In the subsequent phases, we will delve deeper into each component, ensuring that our journey
in creating this state-of-the-art chatbot is as informative as it is transformative for our digital
presence. Stay tuned for more updates and in-depth discussions on each step of this exciting
journey.

Programstrategyhq.com 19
Other Technical Considerations
The development of GenAI applications is a complex process that requires careful planning and
consideration in various aspects such as infrastructure, API integration, costing, and testing. This
blog post aims to provide a detailed guide for professionals embarking on GenAI application
development, especially for those new to this field.

1. Infrastructure Needs
a. Hardware Requirements
• Processing Power: High-performance GPUs and CPUs are critical for training and
running AI models efficiently.
• Memory and Storage: Adequate RAM and storage space to handle large datasets
and model checkpoints.

b. Software and Platforms

• AI and ML Frameworks: TensorFlow, PyTorch, or other machine learning libraries.
• Cloud Platforms: AWS, Google Cloud, or Azure for scalable and flexible computing
resources.

c. Network Infrastructure
• Bandwidth: High-speed internet connectivity for data transfer and cloud interaction.
• Security: Robust network security protocols to protect sensitive data.

2. API Integration
APIs (Application Programming Interfaces) play a crucial role in GenAI applications. They enable
integration with existing systems and facilitate data exchange. Key considerations include:

• Choosing the Right APIs: Based on functionality, compatibility, and performance.

• API Management: Ensuring efficient handling of API requests and responses.
• Security: Implementing authentication and authorization mechanisms to secure API
endpoints.

3. Costing Considerations
Budgeting is a critical aspect of GenAI application development. Key factors affecting costs are:

Programstrategyhq.com 20
 • Infrastructure Expenses: Hardware, software, and cloud services costs.
• Development Costs: Expenses related to the development team, including salaries and
training.
• Operational Costs: Ongoing costs for maintenance, updates, and support.
• Scaling Costs: Additional costs incurred while scaling the application to accommodate
more users or data.

4. Types of Testing Needed

Testing is essential to ensure the reliability and effectiveness of GenAI applications. Essential
testing types include:

• Unit Testing: Verifying individual components or functions.

• Integration Testing: Ensuring that different components of the application work together
seamlessly.
• Performance Testing: Assessing the application’s performance under various conditions.
• Security Testing: Identifying vulnerabilities and ensuring data protection.
• User Acceptance Testing (UAT): Validating the application with end-users to ensure it
meets their needs and expectations.

Compliance with legal standards and ethical considerations is crucial. This includes:

• Data Privacy Laws: Adhering to GDPR, CCPA, or other relevant regulations.

• Ethical AI Practices: Ensuring fairness, transparency, and accountability in AI
applications.

Developing a GenAI application is a multifaceted process that demands meticulous planning in

infrastructure setup, API integration, cost management, and rigorous testing. By considering
these key areas, developers and project managers can build robust, efficient, and compliant
GenAI applications that meet the evolving needs of users and businesses alike.

Programstrategyhq.com 21
GenAI Common Terms
Transformer
Transformers are a type of neural network architecture that revolutionized the way AI systems
process sequential data, like language. Unlike previous models, they don't require data to be
processed in order, allowing for more efficient training and better handling of long-range
dependencies in text. Transformer overcame the limits of RNNs, being able to get much longer
sequences of text as input. Transformers are based on the attention mechanism.

Transformers are fundamental to the development of sophisticated GenAI models, particularly

LLMs. They enable the creation of highly effective language models capable of generating
coherent and contextually relevant text, a key aspect of Generative AI.

Recurrent Neural Networks (RNNs)

RNNs are neural networks designed for sequential data processing. They excel in tasks where
context is key, such as language modeling and speech recognition, by using internal memory to
maintain context from previous inputs.

Model Parameter: Temperature

In neural network models, 'temperature' controls randomness in predictions. A low temperature
results in more predictable outputs, while a high temperature increases diversity and creativity at
the risk of less coherence.

Attention Mechanism
The attention mechanism is a component of neural networks that allows the model to focus on
different parts of the input data, similar to how human attention works. In language tasks, it helps
the model to pay attention to relevant parts of the text when generating a response.

In Generative AI, especially in LLMs, the attention mechanism enhances the model's ability to
generate relevant and coherent text by effectively managing long-range dependencies in
language.

Programstrategyhq.com 22
Retrieval Augmented Generation
Retrieval Augmented Generation (RAG) combines the capabilities of LLMs with an external
knowledge retrieval step. This approach allows the model to pull in information from external
sources, making the generated content more accurate, informative, and relevant.

RAG represents an advancement in Generative AI, especially in knowledge-intensive applications

like question answering and research assistance, where the ability to reference and incorporate
external data is crucial.

Embedding Models
Embedding models transform high-dimensional data (like text) into lower-dimensional vectors,
capturing semantic relationships. Essential in natural language processing (NLP), they improve
performance in tasks like text classification and machine translation by understanding linguistic
nuances.

Vector Database
Vector databases store and manage data in a format optimized for machine learning operations,
specifically vector representations of data. In these databases, information is encoded as vectors,
which are essentially arrays of numbers representing various features of the data.

In GenAI development, vector databases are crucial for efficiently handling and retrieving the
large amounts of vectorized data used in training and operating generative models. They enable
quick similarity searches and are essential for managing the vast datasets needed for training
sophisticated GenAI applications.

Fine Tuning
Fine tuning is a process in machine learning where a pre-trained model is further trained on a
smaller, specific dataset to adapt it to a particular task or context. This approach leverages the
generic capabilities of the model and tailors it to specific requirements.

In GenAI, fine tuning is vital for adapting large, general-purpose models (like LLMs) to specific
applications or industries. It ensures that the generative outputs are not only high quality but also
relevant to the specific context of use.

Programstrategyhq.com 23
Structured & Unstructured Data
Structured data is highly organized and easily searchable, often stored in relational databases
with defined schemas, like SQL databases. Unstructured data, on the other hand, lacks a
predefined format or structure, examples being text, images, and audio.

In GenAI development, both types of data are significant. Structured data helps in training models
with clear, labeled information, while unstructured data, especially text and images, is essential
for training models to understand and generate human-like content.

Tokens
Tokens are the basic units of data processed by a generative AI model. In the context of text, a
token could be a collection of words or symbols or text. Tokens are central to GenAI, particularly
in natural language processing. They allow models to break down and analyze text data at a
granular level, enabling the generation of text that is coherent, contextually appropriate, and
syntactically correct.

Hugging Face
Hugging Face is a company known for its development of natural language processing (NLP)
technologies and tools. They have contributed significantly to the field with their open-source
libraries and pre-trained models.

In GenAI, Hugging Face’s tools and models are extensively used for training and deploying
generative AI models, especially in the field of NLP. Their resources have made advanced GenAI
technologies more accessible to developers and researchers worldwide.

NLP (Natural Language Processing)

NLP is a branch of AI that focuses on enabling computers to understand, interpret, and generate
human language. In the realm of GenAI, NLP is foundational. It powers the ability of models to
process and generate human-like text, making it essential for applications like chatbots, content
creation tools, and language translation services.

Langchain
Langchain is a framework or library designed to facilitate the integration of language models into
applications, making it easier to build complex language-based AI systems. In GenAI
development, Langchain can be instrumental in bridging the gap between the capabilities of

Programstrategyhq.com 24
language models and practical applications, allowing developers to create more sophisticated
and interactive GenAI applications efficiently.

Hallucination
In the context of AI, hallucination refers to instances where a model generates incorrect,
unrealistic, or nonsensical information, often as a result of being presented with ambiguous inputs
or lacking sufficient training data. Addressing hallucination is crucial in GenAI development to
ensure the reliability and accuracy of the generated content. It involves refining training processes
and model architectures to minimize the occurrence of such errors.

Edge Cases
Edge cases are unusual or extreme conditions that occur outside of normal operating parameters,
often not covered in the general training data of a model. In GenAI, handling edge cases is
important to ensure the robustness and reliability of AI models. This involves training the models
with diverse and comprehensive datasets and incorporating scenarios that test the model's ability
to handle rare or unexpected inputs.

Programstrategyhq.com 25
Part 3: GenAI for Project Managers

GenAI Tools

1. ChatGPT for Project Managers: ChatGPT can assist project managers in various tasks
such as automating routine activities, enhancing team communication, and streamlining
project workflows. For more detailed information, you can visit OpenAI's website.

2. GPT Store Tools for Project Managers: The GPT Store offers tools like Project Advisor,
which guides project managers with practical tools and frameworks for effective project
management, and Project Manager GPT, which provides AI alternatives for project
planning. You can explore these tools on the GPT Store website.
- Project Advisor: Project Advisor on GPT Store
- Project Manager GPT: For more options, explore the GPT Store.

3. Google Workspace SlidesAI.io for Presentation: SlidesAI.io, integrated with Google

Workspace, helps in creating AI-powered presentations. It automates slide creation and
provides design suggestions based on given topics. For more details, visit the SlidesAI.io
page on Google Workspace Marketplace.

4. ChatGPT Assistant Plugin for Jira: This plugin integrates ChatGPT with Jira, a project
management tool, to enhance task automation and intelligent response within the Jira
environment. Details about this plugin can be found on the Atlassian Marketplace.

5. Virtual Agile Coach on GPT Store: The Virtual Agile Coach is an AI tool that provides
guidance on Agile methodologies, aiding project managers in implementing Agile
practices. It's available on the GPT Store Virtual Agile Coach on GPT Store.

6. Quillbot: Quillbot is an AI-powered writing and paraphrasing tool that helps in refining
project documentation and creating effective communication. It can be accessed at
Quillbot's official website

Programstrategyhq.com 26
 7. Minutes of Meeting GenAI Tools in the Market: AI tools for generating or summarizing
meeting minutes can be found on various platforms. These tools automatically capture
key points, decisions, and action items from meetings, aiding project managers in
documentation and follow-ups. Open AI

8. Google Assistant: Google Assistant is a virtual assistant powered by AI, useful for project
managers for scheduling, setting reminders, conducting searches, and managing tasks
through voice commands. More information can be found on the Google Assistant
website. Google Assistant website.

How to create your own GPT

Creating a custom GPT model via OpenAI's platform involves a few key steps as outlined on their
Help Center page:

1. Navigate to the GPT Creation Page: Visit chat.openai.com and select "Create a GPT"
or go to "My GPTs" if you already have existing models.
2. Utilize the GPT Builder: In the "Create" tab, you can interact with the GPT Builder to
outline the purpose and functionality of your new GPT. This can include specifics like the
GPT's role or the tasks it should perform.
3. Configure Your GPT: The "Configure" tab allows you to name your GPT, describe its
functions, and select its capabilities, such as web browsing or image creation.
4. Publish Your GPT: Once configured, you can publish your GPT and make it available to
others.
5. Advanced Settings: There are options for more detailed configurations, including adding
images, providing specific instructions, setting prompt starters, uploading knowledge files,
and enabling new capabilities like web browsing or DALL·E image generation.
6. Custom Actions: You can integrate third-party APIs to enhance your GPT's functionality.

For more detailed information and guidance, you can visit the OpenAI Help Center page here.

Programstrategyhq.com 27
GenAI Tools by Work Area

Custom Assistance Tools

1. Azure Bot Service: A Microsoft service enabling the creation of intelligent, enterprise-
grade bots, integrating with various Microsoft services. It features natural language
processing, user intent recognition, and conversational AI, useful for automating
interactions in project management.
• URL: Azure Bot Service
2. Dialogflow: An AI-powered conversational platform by Google for designing and
integrating conversational interfaces into websites, apps, and IoT devices. Known for
natural language understanding, it's useful for stakeholder and team engagement in Agile
environments.
• URL: Dialogflow

Virtual AI Assistants
1. PMOtto: An AI-driven assistant designed for project management, featuring task
scheduling, risk assessment, and stakeholder communication. It integrates with existing
project management software to enhance efficiency.
• URL: PMOtto
2. Lili.ai: Tailored for project management with a focus on risk management, decision-
making, and stakeholder engagement. It adapts to specific project needs and is
particularly useful in Agile environments.
• URL: Lili.ai

Calendar Planning
1. Aurora: An AI scheduling tool ideal for complex project planning, offering resource
optimization, scenario analysis, and conflict resolution. Suitable for managing multiple,
overlapping projects.
• URL: Aurora
2. Liquid Planner: A dynamic project management tool with predictive scheduling, resource
management, and time tracking. Its adaptability makes it a strong fit for Agile
methodologies.
• URL: Liquid Planner

Programstrategyhq.com 28
Agile Tools
1. Pinestem: Focused on Agile methodologies, offering backlog grooming, sprint planning,
and performance metrics. It helps Scrum teams in tracking progress and adapting to
changes.
• URL: Pinestem
2. Tara.ai: An AI-powered platform for streamlining software development. Features include
release planning, progress tracking, and requirement gathering, enhancing efficiency in
Agile projects.
• URL: Tara.ai

Minutes of Meeting Tools

1. Otter: Transcribes and shares meeting notes with real-time transcription, speaker
identification, and keyword searchability. It's valuable for documenting and retrieving
meeting information.
• URL: Otter
2. MeetGeek: An AI assistant for meeting management, providing transcription, summaries,
and action item extraction, capturing key meeting points and tasks.
• URL: MeetGeek
3. Avoma: An AI meeting assistant offering services from pre-meeting preparation to post-
meeting follow-ups, including transcription, summarization, and actionable insights.
• URL: Avoma
4. Supernormal: Turns meetings into searchable, shareable transcripts with auto-generated
summaries, action items, and integrations with other project management tools.
• URL: Supernormal

Each of these tools brings unique capabilities to enhance efficiency, communication, and
decision-making in program management and Agile methodologies, particularly beneficial for
software scrum teams.

Programstrategyhq.com 29
Prompt Engineering

Introduction to Prompt Engineering

This skill, particularly relevant in the context of AI models like OpenAI's GPT (Generative Pre-
trained Transformer), involves crafting inputs (prompts) to these models in a way that maximizes
the quality and relevance of their outputs.

Prompt Engineering is the art and science of designing and refining the inputs given to an AI
model to obtain the most effective and accurate outputs. It's not just about asking questions; it's
about how you ask them. The prompt can be a question, a statement, or a set of instructions, and
the engineer's task is to frame it in a way that guides the AI to produce the desired response.

Prompt design primarily focuses on the creation of specific inputs or questions tailored to elicit
desired responses from an AI model. It's about crafting the content of the prompt itself.

On the other hand, prompt engineering encompasses a broader scope, involving not only the
design of the prompts but also the strategic optimization and refinement of these prompts based
on the model's responses. It includes iterative testing and tweaking to achieve the most accurate
and relevant outputs from the AI, thus integrating both creative and analytical skills.

Prompt engineering is a more comprehensive approach, considering the interaction between

prompt, model, and output in its entirety.

Why is it Relevant to Project Managers?

As a program manager in a software product company, understanding prompt engineering is
crucial for several reasons:

1. Effective Communication with AI Tools: Many project management tasks, such as risk
analysis, resource allocation, and progress tracking, can be augmented by AI. Prompt
engineering skills ensure that you extract the most accurate and relevant information from
AI tools.
2. Enhancing Productivity: By mastering prompt engineering, you can better integrate AI
into your team's workflow, thereby enhancing productivity and efficiency.

Programstrategyhq.com 30
 3. Innovative Problem-Solving: AI can offer novel solutions to complex problems. Knowing
how to query AI systems effectively can unlock innovative approaches to challenges in
project management.

Practical Examples for Project Managers

1. Risk Analysis:
• Poor Prompt: "List project risks."
• Improved Prompt: "Identify potential risks for a software development project
following Agile methodologies, considering factors like team size, technology
stack, and client requirements."
The improved prompt is specific and contextual, leading to more tailored and useful risk
assessments.

2. Resource Allocation:
• Poor Prompt: "How should I allocate resources?"
• Improved Prompt: "Suggest an optimal resource allocation strategy for a cross-
functional team of 10 working on a mid-sized e-commerce application, considering
current workload and individual expertise."
This prompt provides context and specifics, making the AI's suggestions more applicable.

3. Progress Tracking:
• Poor Prompt: "Track project progress."
• Improved Prompt: "Generate a weekly progress tracking template for a Scrum
team focusing on user interface development, including key milestones and
deliverables."
The detailed prompt leads to a more relevant and actionable output.

Prompting Techniques
Key prompting techniques in machine learning, especially for language models like GPT:

1. Chain of Thought Prompting:

• Involves crafting prompts that include a step-by-step reasoning process.

Programstrategyhq.com 31
 • Guides the AI to follow a similar logical pathway in its response.
• Particularly useful for complex problem-solving or providing detailed explanations.

2. Zero-Shot Learning:
• Prompts are designed without providing any examples to the model.
• Enables the AI to understand and execute tasks it has not explicitly been trained
on.
• Useful for testing the model's innate abilities or when specific examples are
unavailable.

3. Few-Shot Learning:
• Involves giving the model a few examples in the prompt to demonstrate the desired
task.
• Helps the AI to quickly adapt to and perform new tasks with minimal input.
• Effective in situations where similar tasks have been encountered but specific
guidance is needed.

4. Decomposed Prompting:
• Breaking down the original prompt into sub-prompts and combining results.
• Useful for multi-hop QA retrieval.
• Specific sub-prompts based on a few-shot prompt for the use case

Prompt Engineering is a critical aspect of effective communication in the age of AI. As AI continues
to permeate various aspects of business and technology, the ability to interact efficiently and
effectively with these systems will become increasingly important.

Programstrategyhq.com 32
Prompt Generation Tools
There are several tools available that can assist in generating prompts for AI applications, each
with its own set of features and advantages. Here's a summary of some notable ones:

1. PromptGen: Offers a simple user interface and quick response time. It's free for basic
use, but the Pro version is available at $10/month. This tool is suitable for generating basic
genre filters and keyword input prompts.
2. Taskade: Known for its intuitive UI and multiple output formats like lists, boards, and mind
maps. It also features real-time collaboration. The basic version is free, while the Pro starts
at $7/month. It's particularly useful for generating prompts for a variety of purposes
including marketing, branding, product development, and more.
3. PromptPerfect: Specializes in automatic prompt optimization for AI text and image
models. It caters to non-native English speakers with prompt enrichment and translation
features, and integrates with various platforms like ChatGPT and WordPress. The free
plan includes 20 credits, with a paid plan starting at $9.99 per month.
4. PromptoMANIA: A free AI prompt builder supporting popular AI image generator models.
It's ideal for creating detailed art and visuals with various customization options. It's
particularly beneficial for graphic designers and illustrators.
5. WebUtility: A random AI prompt generator that is effective for generating prompts across
various AI models. It's a great tool for overcoming creative blocks and is highly versatile,
suitable for designers, developers, and business owners.

Programstrategyhq.com 33
 6. PromptBase: Offers a wide range of customizable prompts available for purchase,
catering to various AI engines. It's known for its high-quality prompts and has been
featured in reputable publications.

Additional reading: Prompt Design and Engineering: Introduction and Advanced Methods by
Xavier Amatriain - Cornell University Arxiv

Programstrategyhq.com 34
GenAI Use cases in Project Management

1. Automated Content Creation: Enhancing efficiency and consistency in producing

marketing materials, product descriptions, and customer communications.
2. Coding Assistance: Supporting software development by generating code from natural
language descriptions and maintaining code across different platforms.
3. Advanced Analytics: Facilitating complex data analysis and generating insights, helping
in strategic decision-making.
4. Customer Interaction: Enhancing customer support with AI-powered virtual agents that
provide real-time, personalized assistance.
5. Process Optimization: Streamlining business processes by automating repetitive tasks
and improving operational efficiency.
6. Innovation Acceleration: Aiding in rapid prototyping and product development by
generating creative concepts and designs.
7. Market Research Enhancement: Utilizing AI for in-depth market analysis, identifying
trends, and generating customer insights.

Deloitte Consulting LLP, 2023. Generative AI Use Cases. [online] Available at:
https://www2.deloitte.com/content/dam/Deloitte/us/Documents/consulting/us-ai-institute-gen-ai-
use-cases.pdf [Accessed 28 January 2024].

Use cases for Project Management Bots

1. Task Creation and Prioritization Agent: Automates task generation and prioritizes them
based on project needs and deadlines.
2. Meeting Scheduling Agent: Streamlines meeting planning by finding optimal times and
managing invites.
3. Project Status Reporting Agent: Generates real-time reports on project progress and
status.
4. Document Sorting and Management Agent: Organizes project-related documents for
easy access and retrieval.
5. Project Planning and Scheduling Agent: Assists in creating detailed project plans and
schedules.

Programstrategyhq.com 35
 6. Task Automation Agent: Automates routine tasks to increase efficiency.
7. Agile Project Management Agent: Supports Agile methodologies with tools for sprint
planning and tracking.
8. Risk Management Agent: Identifies potential project risks and suggests mitigation
strategies.
9. Risk Assessment Generator: Simplifies the process of identifying and evaluating project
risks.
10. Project Planning Generator: Streamlines creating comprehensive project plans.
11. Scope Definition Generator: Aids in defining clear project boundaries and deliverables.
12. Timeline Creation Generator: Facilitates the development of detailed project timelines.
13. Requirements Generation Tool: Helps in articulating project needs and specifications.
14. Issue Management Tool: Streamlines the process of identifying and addressing project
issues.
15. Stakeholder Analysis Tool: Assists in identifying and analyzing project stakeholders.
16. Project Charter Creation Tool: Simplifies the development of a formal project charter.
17. Resource Allocation Tool: Enhances the process of distributing resources effectively
across a project.

Taskade. (n.d.). AI Project Management GPTs and Agents. Available at:

https://www.taskade.com/agents/project-management [Accessed 28 Jan. 2024].

Use Cases Generator

The GenAI Usecases Generator is a specialized GPT variant tailored for project management
professionals. Its main objective is to provide innovative and actionable ideas for GenAI
applications in the field of project management. Here's an overview and a use case example:

1. Description and Purpose:

• The GenAI Usecases Generator is designed to suggest 2 to 3 detailed options per
query.
• Its focus is on applications that enhance project management through automation,
assistance, or augmentation.
• It includes technical details like suitable Large Language Models (LLMs) and
development steps, ensuring practical and implementable suggestions.

Programstrategyhq.com 36
 2. Usage for Project Managers
• Project managers can leverage this GPT to develop new tools or plugins that utilize
AI to streamline project management processes.
• It assists in enhancing team collaboration and automating routine tasks.
• The GPT can be consulted for solutions to specific project management
challenges, offering tailored advice based on the latest AI technologies.

3. Example Use Case

• A project manager is struggling with maintaining efficient communication within a
globally distributed team.
• The GenAI Usecases Generator might suggest the creation of an AI-powered
communication assistant.
• Key features of this tool would include:
• Automatic scheduling of meetings considering all time zones.
• Summarizing key points from asynchronous discussions.
• Real-time translation to facilitate communication across different
languages.
• The development of this tool would involve:
• Using OpenAI's language models.
• Integrating with platforms like Slack or Microsoft Teams.
• Customizing the tool to fit the specific communication patterns of the team.

For more information and to access this service, project managers can visit GenAI Usecases
Generator.

This specialized GPT acts as an enabler for project management needs to develop AI-driven
solutions, offering practical, innovative, and customized advice to enhance GenAI adoption.

Programstrategyhq.com 37
How to Plan Gen AI Projects

Discover how Generative AI projects need to be planned effectively to get the desired outcomes.
This insightful blog post delves into the integration of AI tools and methodologies, offering a
perspective on managing GenAI development projects. It highlights the challenges, opportunities,
and practical implications of incorporating agility and design thinking into project planning and
execution. Ideal for program managers and Scrum teams looking to innovate and streamline their
processes.

For an in-depth exploration of this topic, visit: GenAI Project Planning on Program Strategy HQ.

Programstrategyhq.com 38
Gen AI Project Metrics

As organizations strive to maximize their AI investments, the challenge lies not only in deploying
these technologies but also in measuring their success and aligning them with business goals.

The Role of KPIs in Generative AI Implementation

KPIs serve as a compass in navigating the complex terrain of AI implementations. They are
instrumental for:

1. Assessing Performance Objectively: By providing quantifiable measures, KPIs help in

evaluating the efficacy of AI projects.
2. Aligning with Business Goals: KPIs ensure that AI initiatives are in sync with the broader
objectives of the organization.
3. Facilitating Data-Driven Adjustments: Regular KPI assessments allow for timely
modifications, enhancing the AI system's performance.
4. Enhancing Adaptability: They provide insights that help in adapting AI strategies to
evolving business needs.
5. Stakeholder Communication: Clear KPIs aid in transparent reporting to stakeholders,
elucidating the progress and impact of AI projects.
6. Demonstrating ROI: They are crucial in proving the return on investment of AI initiatives,
justifying further expenditure and development.

Key Areas of Focus for KPIs in Generative AI

Model Quality
• Quality Index: An aggregate of various metrics to represent overall model
performance (e.g., BLEU, Rouge).
• Error Rate: The percentage of incorrect or invalid responses, determined through
human evaluation.
• Latency: The delay between a query submission and response, factoring in model
architecture and infrastructure.
• Accuracy Range: A predefined threshold of precision that the model should meet.

Programstrategyhq.com 39
 • Safety Score: Assessing the model's handling of sensitive topics relevant to the
business.

System Quality
• Data Relevance: Ensuring all data is pertinent to the model, avoiding biases and
inefficiencies.
• Data and AI Asset Reusability: The extent to which data and AI assets are
discoverable and reusable.
• Throughput: The model's capacity to handle data volume within a specific
timeframe.
• System Latency: Total response time, including network, data, and model
latencies.
• Integration and Backward Compatibility: The ability to integrate with existing
systems and consider future model upgrades.

Business Impact
• Adoption Rate: The ratio of active users to the total intended audience.
• Frequency of Use: Regularity of queries submitted by users.
• Session Length: Average duration of user interactions.
• Queries per Session: Number of queries per user interaction.
• Query Length: Average size of user queries.
• Abandonment Rate: Percentage of sessions ending before fulfilling user queries.
• User Satisfaction: Measured through surveys and metrics like Net Promoter
Score (NPS).

As generative AI continues to revolutionize various sectors, the importance of KPIs cannot be

overstated. They are not just metrics but strategic tools that guide organizations in harnessing the
full potential of AI technologies. By meticulously tracking and analyzing these KPIs, businesses
can ensure that their AI investments are not just innovative but also aligned with their overarching
goals and yielding tangible benefits.

Google Cloud. (n.d.). KPIs for gen AI: Why measuring your new AI is essential to its success.
Retrieved from https://cloud.google.com/transform/kpis-for-gen-ai-why-measuring-your-new-ai-
is-essential-to-its-success.

Programstrategyhq.com 40
Resources – Gen AI Online Courses & Books
Online Video Courses

1. Google Cloud Skills Boost - Introduction to Generative AI Learning Path

This learning path, managed by Google Cloud, offers an introductory insight into
generative AI. It encompasses various courses, including fundamentals of large language
models, an overview of generative AI, and responsible AI principles. The path aims to
enhance understanding and practical application of AI concepts in real-world scenarios.
Url - Generative AI Learning Path

2. Harvard University's Data Science: Machine Learning Course on edX

This online course covers key concepts such as training data, cross-validation, and
various machine learning algorithms. The course, designed to be self-paced over 8 weeks
with 2-4 hours of weekly commitment, is part of Harvard's Professional Certificate Program
in Data Science. It's ideal for learners seeking to understand machine learning's role in
data science, with emphasis on practical application through a project-based approach.
Participants can choose to audit the course for free or opt for a verified certificate for $149.
For more details visit Data Science: Machine Learning

3. Duke University's AI Product Management Specialization on Coursera

This specialized program, offered by Duke University on Coursera, provides a
comprehensive introduction to AI Product Management. It is designed to impart
foundational knowledge about how machine learning works and its practical applications.
The specialization is structured to be accessible to professionals from various industries
and functions, including product managers and engineering team leadersThe
specialization is beneficial for professionals aiming to understand the application of AI in
product management and for those seeking to lead cross-functional teams in machine
learning projects. For more information, you can visit AI Product Management

4. Foundational Generative AI – Video Series

A dynamic learning environment in the fascinating topic of generative artificial intelligence
is provided by the Generative AI Community Edition course. This course, which focuses
on community participation, offers thorough insights into generative models and different
frameworks, investigating the production of text and visual material produced by AI. Link
to the Course Foundational Generative AI

Programstrategyhq.com 41
5. Generative AI for Beginners
Generative AI for Beginners"is an educational resource hosted on GitHub by Microsoft.
This course offers a comprehensive introduction to Generative AI, targeting beginners in
the field. It covers fundamental concepts and applications of Generative AI, providing a
solid foundation for those new to this area of artificial intelligence. The course is designed
to be accessible and engaging, making it a valuable learning tool for individuals seeking
to understand and apply Generative AI in various contexts. You can explore the course in
detail here - A 12 Lesson course teaching everything you need to know to start building
Generative AI applications

6. AI, Machine Learning & Chatbots for Product Managers

A video playlist from Product School Link to the series

Generative AI Top books

1. Generative Deep Learning: Teaching Machines to Paint, Write, Compose, and Play by
David Foster
This book provides a practical introduction to generative deep learning, offering step-by-
step guidance on creating models that can generate original content. It's suitable for
readers who have some understanding of machine learning and are interested in exploring
creative applications of AI.

2. Deep Learning by Ian Goodfellow, Yoshua Bengio, and Aaron Courville

While not solely focused on generative AI, this book is a foundational text in deep learning.
It covers a range of topics including generative models, making it a must-read for anyone
serious about understanding the underlying principles of AI technologies.

3. AI Superpowers: China, Silicon Valley, and the New World Order" by Kai-Fu Lee
This book offers a compelling look at the future of global technology, focusing on the race
in AI between China and the United States. Written by Kai-Fu Lee, a prominent figure in
the global AI community, it provides insights into how AI will shape the future of business
and the global economy. It's particularly suitable for business leaders and managers who
want to understand the broader impact of AI technologies like generative AI on
international business and industry trends.

Programstrategyhq.com 42
4. Competing in the Age of AI: Strategy and Leadership When Algorithms and Networks
Run the World" by Marco Iansiti and Karim R. Lakhani
This book is specifically tailored for business leaders and managers. It explores how AI is
transforming business models and offers a framework for thinking about AI strategy in
organizations. The authors, both of whom are Harvard Business School professors,
provide insights into how companies can leverage AI technologies, including generative
AI, for competitive advantage.

5. The Hundred-Page Machine Learning Book by Andriy Burkov

Though not exclusively about generative AI, this book offers a concise overview of key
machine learning concepts, including generative models. It's an excellent quick-reference
guide for understanding the broader context in which generative AI operates.

Thank you and Best Wishes on your GenAI Journey!

Programstrategyhq.com 43
Generative AI for Managers at PSHQ

Follow our website GenAI section for more learning and opportunities.

Reach us at [email protected]

Programstrategyhq.com 44
References

1. Microsoft Azure. (n.d.). Machine Learning. Retrieved from

https://azure.microsoft.com/en-us/products/machine-learning/
2. Microsoft. (n.d.). Generative AI for Beginners. Retrieved from
https://microsoft.github.io/generative-ai-for-beginners/#/
3. Xavier Amatriain. (2024). Prompt Design and Engineering: Introduction and Advanced
Methods. arXiv. Retrieved from https://arxiv.org/abs/2401.14423.
4. ProgramStrategyHQ. (n.d.). GenAI Project Planning. Retrieved from
https://www.programstrategyhq.com/post/genai-project-planning
5. Taskade. (n.d.). Project Management. Retrieved from
https://www.taskade.com/agents/project-management
6. Henrik Kniberg (2024) . Generative AI in a Nutshell - how to survive and thrive in the age
of AIYouTube. Retrieved from https://www.youtube.com/watch?v=2IK3DFHRFfw
7. Deloitte. (n.d.). Generative AI use cases. Deloitte. Retrieved from
https://www2.deloitte.com/content/dam/Deloitte/us/Documents/consulting/us-ai-institute-
gen-ai-use-cases.pdf
8. Google Cloud. (n.d.). KPIs for Gen AI: Why measuring your new AI is essential to its
success. Retrieved from https://cloud.google.com/transform/kpis-for-gen-ai-why-
measuring-your-new-ai-is-essential-to-its-success.
9. Wikipedia contributors. (n.d.). Generative artificial intelligence. Wikipedia. Retrieved from
https://en.wikipedia.org/wiki/Generative_artificial_intelligence
10. Google Cloud. (n.d.). Generative AI. Retrieved from https://cloud.google.com/use-
cases/generative-ai
11. McKinsey & Company. (n.d.). What is Generative AI? Retrieved from
https://www.mckinsey.com/featured-insights/mckinsey-explainers/what-is-generative-ai
12. Techopedia. (n.d.). Generative AI. Retrieved from
https://www.techopedia.com/definition/34633/generative-ai

Programstrategyhq.com 45