DEVELOPING GE
NERATIVE AI SOL
UTIONS
MODULE -4
LEARNING
OUTCOME
• Understand Prompt
Engineering
• Apply Prompt Engineering
Techniques
• Interact with Prompt Labs
• Understand Foundation Model
Parameters
• Improve Output Quality
• Ground Outputs in Factual
Data
TODAYS OBJECTIVE
• Understand what prompt
engineering is
• Explore types and formats of
prompts
• Learn techniques for writing
effective prompts
• Understand prompt impact on
output
WHAT IS PROMPT
ENGINEERING?
Prompt Engineering is the
practice of designing and
optimizing input prompts to
eff ectively interact with AI
language models to get the
best possible output for a given
task.
• Example: Like giving clear instructions to a smart assistant
• Better prompts = better results
WHAT IS
ARTIFICIAL
INTELLIGENCE?
Artificial Intelligence is not just a
buzzword; it's a transformative force that
is reshaping industries, solving complex
problems, and revolutionizing the way we
live and work.
So, let's dive into the world of Artificial
Intelligence, where machines become
intelligent companions in our quest for
progress and innovation.
WHY PROMPT
ENGINEERING
MATTERS?
Unlocks Full Potential of AI Models
Improves Accuracy and Reliability
Enhances Reasoning and Creativity
Critical for Business and
Professional Use
Saves Time and Effort
Drives ROI in AI Investments
Essential for Building AI Products
1. UNLOCKS FULL POTENTIAL
OF AI MODELS
Large language models (LLMs) like
GPT-4 or IBM watsonx.ai are powerful,
but without the right prompt, they:
• May misunderstand the task
• Produce irrelevant or verbose
results
• Fail to follow critical instructions
2. IMPROVES ACCURACY AND
RELIABILITY
Poor prompts lead to poor output. A well-engineered
prompt can:
• Minimize hallucinations (wrong or made-up facts)
• Produce more contextually accurate and trustworthy
answers
• Enhance factual correctness and adherence to
constraints
3. ENHANCES REASONING
AND CREATIVITY
With techniques like Chain-of-Thought prompting, you
can guide the AI to:
• Think step-by-step
• Justify answers
• Be more creative or analytical, depending on the task
SAVES TIME AND EFFORT
Rather than coding complex pipelines or workflows:
• Prompt engineering lets you solve tasks with plain
language
• It accelerates prototyping and content generation
Example: Generate legal summaries, analyze resumes,
write code—all with a few words.
5. CRITICAL FOR BUSINESS AND
PROFESSIONAL USE
In enterprise or professional settings, AI must:
• Follow tone, structure, and compliance guidelines
• Provide outputs in usable formats (JSON, table, markdown)
Prompt engineering enables that control—making AI a reliable
tool in:
• Customer support
• Content creation
6. DRIVES ROI IN AI INVESTMENTS
Organizations adopting AI need:
• High-quality, consistent results
• Reusable prompt templates
• Scalable and explainable outputs
Eff ective prompting leads to fewer errors, better adoption, and
faster deployment.
7. ESSENTIAL FOR BUILDING AI
PRODUCTS
Whether you're building:
• A chatbot
• An AI tutor
• A code assistant
• A marketing content generator
...prompt engineering is the core design skill that shapes the
product's capabilities.
WHERE IS
PROMPT Code
ENGINEERING generation
USED?
Data Creative
summarizatio writing
n
Chatbots & Customer
Virtual support
Assistants automation
Education
and training
tools
TYPES OF
PROMPTS
• Zero-shot: No example provided
• One-shot: One example given
• Few-shot: Multiple examples provided
• Instructional: Explicit command (e.g., “Summarize
this…”)
• Conversational: Maintains dialogue context
PROMPT
ENGINEERING
TECHNIQUES
• Use clear instructions
• Specify format (“Output in JSON”, “Bullet
points”)
• Add context (e.g., task role: “You are a
legal assistant…”)
• Use examples to guide behavior
• Apply constraints (e.g., “max 50 words”)
TYPES OF MACHINE
LEARNING?
Supervised Learning Unsupervised Learning
Reinforcement Learning
PROMPTING
EXAMPLES
• Summarize this text in 3 bullet
points
• Write a Python function to sort a
list.
• Classify sentiment of the
review: <text>
• Extract name and date from:
<text>
PROMPT LAB IN IBM
WATSONX.AI
CODE
GENERATION
THANK
YOU
ANY
QUERIES
