Prompt Engineering - Content
Creation
Export of https://partyrock.aws/u/js2222/twzxogwys/Prompt-Engineering-Content-Creation
Overview
Welcome to Content Creation
In this chapter, we will learn how to write prompts to generate content. Use cases include:
● Brainstorming
● Outlining
● First draft
This chapter is part of the Prompt Engineering Guide
Before you start these exercises, keep in mind:
1: For all exercises, copy the text from the example box and paste it into the neighboring
prompt box. The results from the large language models will be displayed in the
corresponding response boxes.
2: The response boxes indicate which large language model is used to respond (AI21
Jurassic, Amazon Titan, Anthropic Claude, or Cohere Command). You will see different
models used throughout the guide.
3: Compare the responses between the models. Think about why they might have
responded differently.
4: Think about how you could rewrite the prompts to change the responses to suit your
intended purpose.
Exercise 1
1. Brainstorming
Blocked on where to start? Have the model generate a list of ideas for you!
Copy example below and paste into Prompt Box 1
Please create a list of startup ideas for a new generative AI service for the food industry:
Prompt Box 1
Response 1 using Claude Sonnet
Response 1 using Command
Response 1 using Jurassic Ultra
Challenge 1
Can you alter the prompt above to add more details for each idea?
Exercise 2
2. Brainstorming with more details
You can also influence the format of responses by letting the model know what you want to
see:
Copy example below and paste into Prompt Box 2
Please create a list of startup ideas for a new generative AI service for the food industry. For
each startup idea, provide a name and a one-sentence elevator pitch for what the startup will
do.
Prompt Box 2
Response 2 using Command
Response 2 using Jurassic Ultra
Response 2 using Claude Sonnet
Challenge 2
Can you alter the prompt to be more specific, or add more details?
Could you have it focus on restaurants, food manufacturers, or consumers?
Can you have it indicate the target market for each startup idea?
Exercise 3
3. Slogan
You can have the model generate short content like a marketing slogan, based on existing
material:
Copy example below and paste into Prompt Box 3
Amazon Bedrock is a fully managed service that offers a choice of high-performing
foundation models (FMs) from leading AI companies like AI21 Labs, Anthropic, Cohere,
Meta, Stability AI, and Amazon with a single API, along with a broad set of capabilities you
need to build generative AI applications, simplifying development while maintaining privacy
and security. With the comprehensive capabilities of Amazon Bedrock, you can easily
experiment with a variety of top FMs, privately customize them with your data using
techniques such as fine-tuning and retrieval augmented generation (RAG), and create
managed agents that execute complex business tasks—from booking travel and processing
insurance claims to creating ad campaigns and managing inventory—all without writing any
code. Since Amazon Bedrock is serverless, you don't have to manage any infrastructure,
and you can securely integrate and deploy generative AI capabilities into your applications
using the AWS services you are already familiar with.
Create a slogan for the above product:
Prompt Box 3
Amazon Bedrock is a fully managed service that offers a choice of high-performing
foundation models (FMs) from leading AI companies like AI21 Labs, Anthropic, Cohere,
Meta, Stability AI, and Amazon with a single API, along with a broad set of capabilities you
need to build generative AI applications, simplifying development while maintaining privacy
and security. With the comprehensive capabilities of Amazon Bedrock, you can easily
experiment with a variety of top FMs, privately customize them with your data using
techniques such as fine-tuning and retrieval augmented generation (RAG), and create
managed agents that execute complex business tasks—from booking travel and processing
insurance claims to creating ad campaigns and managing inventory—all without writing any
code. Since Amazon Bedrock is serverless, you don't have to manage any infrastructure,
and you can securely integrate and deploy generative AI capabilities into your applications
using the AWS services you are already familiar with.
Create a slogan for the above product:
Response 3 using Claude Sonnet
Here are a few potential slogans for Amazon Bedrock:
1. "Bedrock: The Foundation for Your AI Future"
2. "Unleash AI Potential with Amazon Bedrock"
3. "Build Smarter, Faster with Bedrock"
4. "Amazon Bedrock: Where AI Meets Simplicity"
5. "Elevate Your AI Game with Amazon Bedrock"
6. "Bedrock: Solid Ground for Limitless AI Innovation"
7. "Amazon Bedrock: Powering the AI Revolution"
8. "Transform Your Business with Bedrock's AI Arsenal"
9. "Bedrock: The Cornerstone of Modern AI Development"
10. "Simplify AI, Amplify Results with Amazon Bedrock"
Response 3 using Command
"Amazon Bedrock: Powering Your AI Applications Privately and Securely"
Response 3 using Jurassic Ultra
"Amazon Bedrock: Simplifying Generative AI with Unmatched Flexibility and Security."
Challenge 3
Can you come up with some new slogans for other products?
Exercise 4
4. Rewriting from bullet points
The models can be used to take your raw thoughts in bullet point forms and generate
content from them:
Copy example below and paste into Prompt Box 4
Use cases for generative AI:
* Summarizing large documents
* Answering questions based on a body of knowledge
* Analyzing content and providing recommendations
* Creating new content
Please write an introductory paragraph to a blog article highlighting the above use cases:
Prompt Box 4
Use cases for generative AI:
● Summarizing large documents
● Answering questions based on a body of knowledge
● Analyzing content and providing recommendations
● Creating new content
Please write an introductory paragraph to a blog article highlighting the above use cases:
Response 4 using Jurassic Ultra
Response 4 using Claude Sonnet
Generative AI is revolutionizing the way we interact with and process information, offering a
wide range of applications that can significantly enhance productivity and creativity across
various industries. From summarizing lengthy documents to providing intelligent answers
based on vast knowledge bases, this cutting-edge technology is proving to be an invaluable
tool for businesses and individuals alike. Moreover, generative AI's ability to analyze content
and offer tailored recommendations is transforming decision-making processes, while its
capacity to create original content opens up new possibilities for creative endeavors. In this
article, we'll explore four key use cases of generative AI that are reshaping the way we work
and innovate: summarizing large documents, answering questions based on a body of
knowledge, analyzing content and providing recommendations, and creating new content.
Response 4 using Command
Generative AI is transforming the way we approach various tasks, unlocking new
possibilities and enhancing our efficiency. With its incredible capabilities, this technology can
process vast amounts of information and generate valuable insights, saving us valuable time
and effort. Whether it's summarizing extensive documents, answering complex questions,
analyzing content and providing tailored recommendations, or even creating fresh and
innovative content, generative AI has got you covered. In today's fast-paced world, where
time is precious and information overload is a daily struggle, these use cases showcase how
generative AI is a game-changer, revolutionizing the way we work and live. Stay tuned to
explore each of these applications further and discover the potential of this exciting
technology.
Challenge 4
Could you alter the prompt above to produce a social media post instead?
Congratulations
You have completed the Content Creation chapter!
Try another chapter in the Prompt Engineering Guide:
1: Prompt Engineering Introduction
2: Content Creation
3: Providing Context to Reduce Hallucinations
4: Targeting Your Questions
5: Repurposing Content
6: Summarizing Long Content
7: Language Translation
8: Analyzing Customer Communication
9: Information Extraction
10: Prompting to Create and Understand Code
11: Doing More with Generative AI
Find more generative AI labs at the Building with Amazon Bedrock and LangChain
workshop.
