DOWNLOAD COURSE MIND MAP FROM HERE.

Generative AI Solution Development

From Prompt Engineering to RAG

Prompt Engineering Primer

Prompt → an input or query given to an LLM to elicit a specific response or output.

Prompt engineering → practice of designing and refining prompts to optimize the responses generated by AI models.

A good prompt usually consists of:

• Instruction
• Context
• Input / question
• Output type / format

Prompt Engineering Techniques

Zero-shot Prompting → no examples are provided

Few-shot Prompting → a few input-output examples are provided to guide the model

Prompt Chaining → break tasks into subtasks:

• Multiple tasks are linked together, with the output of one prompt serving as the input for the next.
• This method allows for more complex tasks to be broken down into manageable steps.

Chain-of-Thought Prompting → “let’s think step by step”:

• Chain-of-Though (CoT) prompting enhances the reasoning capabilities of LLMs by guiding them to articulate their thought processes step-by-step, similar to human reasoning.
• Research in this area has mixed findings.

Prompt Engineering Tips & Tricks