Duration
2-days
Skills Gained
By the end of this course, participants will be able to:
-
Apply advanced prompt engineering techniques to maximize GenAI results
-
Apply chain-of-thought and step-based prompting to complex problems
-
Apply meta-prompting to direct specific outputs and improve AI performance
-
Improve agentic AI capabilities through specific prompting strategies
-
Apply ensembling, self-consistency, and self-criticism to push past basic prompting
Who Can Benefit
This course is designed for:
-
Anyone interested in learning prompt engineering, including business users, managers, data analysts, data scientists, developers, and more
Prerequisites
Participants should enter this course with:
Organizational Objectives
This course assists organizations to:
-
Reduce time on first-draft work by codifying high-quality prompt patterns across teams
-
Lower model-run cost by reaching for prompt engineering before fine-tuning
-
Build a working knowledge of agentic prompt patterns across the organization
-
Establish a shared prompt-quality vocabulary across functions
Software
All attendees must have a modern web browser and an Internet connection.
Module 1 - Understanding Generative AI
By the end of this module, you will be able to differentiate generative AI from traditional AI, describe how foundation models produce text and other modalities, and recognize the model settings and parameters that influence output quality.
-
Defining intelligence, artificial intelligence, and generative AI
-
Differentiating generative AI from traditional AI
-
How generative AI works (tokenization, embeddings, decoding)
-
Benefits and challenges of generative AI
-
Popular GenAI models and frameworks
-
LLM settings and parameters; intro to prompt engineering, RAG, and agentic AI
-
Hands-on Lab: Explore tokenization, embeddings, and core LLM settings hands-on without writing any code.
Module 2 - Prompting with Examples
By the end of this module, you will be able to apply zero-shot and few-shot prompting, recognize the limitations of N-shot examples, and use dynamic example selection to keep prompts efficient as datasets grow.
-
Introduction to prompting with examples
-
Zero-shot prompting
-
Few-shot prompting
-
Limitations of few-shot prompting
-
Best practices for prompting with examples
-
Dynamic examples
-
Hands-on Lab: Implement N-shot prompting with dynamic example selection on one realistic task and measure the quality lift.
Module 3 - Chain-of-Thought and Step-Based Prompting
By the end of this module, you will be able to apply chain-of-thought, contrastive CoT, auto-CoT, and step-back prompting to multi-step reasoning tasks, and recognize emergent prompting techniques like Logic-of-Thought and Narrative-of-Thought
-
Introduction to chain-of-thought prompting
-
Zero-shot chain-of-thought
-
Contrastive chain-of-thought
-
Auto-chain-of-thought
-
Step-back prompting
-
Logic-of-thought, Narrative-of-thought, and emergent techniques
-
Hands-on Lab: Compare CoT, zero-shot CoT, and contrastive CoT on the same reasoning task using aligned examples.
Module 4 - Meta-prompting and Reflective Prompting
By the end of this module, you will be able to apply meta-prompting to direct model behaviour at a higher level, distinguish it from chain-of-thought, decompose tasks via prompt chaining, and apply Reflexion-style reflective prompting.
-
Characteristics of meta-prompting
-
Meta-prompting vs. chain-of-thought
-
Benefits and challenges of meta-prompting
-
Prompt chaining and task decomposition
-
Reflexion
-
Hands-on Lab: Implement meta-prompting and Reflexion on a multi-step task and compare quality against single-prompt baselines.
Module 5 - Prompting Agentic AI
By the end of this module, you will be able to apply prompt engineering to control agent tool use, integrate memory and context, structure goal-oriented prompts, and combine prompting techniques into a coherent agent strategy.
-
Core concepts of agentic AI
-
Prompting to control tool use
-
Integrating memory and context
-
Prompting for goal-oriented tasks
-
Integrating prompting techniques for agents
-
Hands-on Lab (choice of track): Build a goal-oriented agent that uses tools and integrates memory and context.
-
No-Code Track (business users, analysts): Assemble the agent's tool-use and memory prompting in a guided, point-and-click builder - no programming required.
-
Developer Track (data scientists, developers): Use LangGraph to control tool use and integrate memory and context into the same agent.
Module 6 - Advanced Prompting Techniques
By the end of this module, you will be able to apply ensembling, code prompting, self-consistency, self-criticism, and plan-and-solve techniques to push prompt-based performance past basic methods.
