Prompt Engineering II
Overview
Prompt engineering is the practice of designing, testing, and refining the input text (or “prompt”) given to a language model or other AI system to produce a desired output. Prompt engineering involves understanding how to communicate clearly and effectively with AI models. Prompt Engineering II continues from our Prompt Engineering I class to introduce students to the core intermediate and advanced prompting techniques.
By the end of this course, participants will be able to:
-
Design and implement complex prompt strategies utilizing advanced techniques
-
Analyze and deconstruct existing sophisticated prompts
-
Develop and apply prompt engineering best practices for specialized AI applications
-
Integrate prompt engineering principles
-
Evaluate and compare the performance of different prompting methodologies
This course is included in the AI Learning Subscription.
Schedule
Register 21 days before class start date and save 10%! Enter discount code EARLY10 during registration.
Dates
Times
Location
Price
GTR: Guaranteed to Run
Course Outline
Prompt Engineering II
1. Review of Intermediate Techniques
- Quick refresher on role-based, iterative, and few-shot prompting.
- Discussion: When to use intermediate techniques.
- Hands-on: Fine-tune prompts for specific outcomes.
2. Advanced Prompting Strategies
- Chain-of-Thought (CoT):
- Step-by-step reasoning for problem-solving.
- Demo: Solve logic puzzles or explain complex processes.
- Tree-of-Thought (ToT):
- Exploring branching possibilities for ideation or decision-making.
- Demo: Brainstorm and evaluate solutions for a community issue.
- Self-Consistency:
- Generating multiple outputs and finding the most consistent answer.
- Demo: Answer trivia questions using self-consistency.
- Practice: Apply CoT, ToT, and self-consistency to real-world tasks.
- Additional Advanced Techniques (75 minutes)
- Meta-Prompting: Critique or improve model outputs.
- Retrieval-Augmented Prompting: Incorporating external data or documents.
- Debate and Critique Prompting: Arguing opposing viewpoints.
- Constraint-Based Prompting: Imposing limits on style, format, or word count.
- Demo:
- Meta-Prompting: “Improve your explanation for a beginner audience.”
- Constraint-Based: Write a 50-word product pitch.
- Hands-on: Create prompts for advanced scenarios.
- Interactive Exercises and Case Studies
- Group task: Use advanced techniques to solve a complex, multi-step problem (e.g., plan an event, write a policy proposal).
Compare outputs and discuss optimization strategies.
3. Wrap-Up and Next Steps (15 minutes)
- Recap advanced concepts.
- Resources for further learning.
- Open Q&A.