Career ColumnsCareer Power-UpCareer Skills

Win Gift Giving this Holiday Season with Prompt Engineering

By Paige Kassalen

Prompt engineering is the process of crafting precise and strategic inputs (prompts) to guide AI models toward generating relevant and useful responses. It’s a must-have skill for anyone working with generative AI models like ChatGPT, Gemini or Claude.

This skill is essential because generative AI models rely on “hidden patterns” within the data they were trained on to help understand complex ideas and connections, creating realistic and meaningful responses. These hidden patterns enable the models to produce useful results without needing users to specify every minute detail for each situation.

For example, when engineering a prompt, you could ask ChatGPT to make an email “friendlier.” The model then uses these hidden patterns, learned from its training data, to make decisions about how to adjust the tone — such as including exclamation marks, a warm greeting, and a positive tone — without requiring the user to outline each of these specifics.

Understanding how generative AI makes these decisions behind the scenes is important, as a well-crafted prompt can “nudge” the model toward certain associations and preferences, helping it to produce the most relevant and desired result.

Now, how does this relate to finding the best gifts?

So many times, I’ve found myself combing through article after article, searching for a unique gift for a friend or family member, only to get exhausted by the same generic suggestions that don’t match what I’m looking for.

Generative AI can help with this! I can ask ChatGPT questions like, “What are the top gifts for men this holiday season?” or “What gift should I get for my dad?” ChatGPT then compiles information based on these prompts and provides a tailored response.

This approach saves me hours of research, and with effective prompt engineering, we can get even better results.

How can you engineer a stronger prompt?

As I mentioned above, generative AI models use “hidden patterns” they learn from training data to infer what the desired output should include. To engineer an effective prompt, be as specific as possible about what you want.

Include details about what the person likes, what they don’t like, whether they prefer experiences over material possessions, your budget, and any other criteria that will help guide you to your desired outcome.

Here’s an example of my prompt when trying to find a gift for my dad:

“I’m seeking 5 novel gift ideas for my 63-year-old father for Christmas. He enjoys Tai Chi and playing the Native American wooden flute. He enjoys biking outside and has multiple bikes, including an electric bike. He recently has been interested in watches, but he does not want a new watch. He does not like to accumulate too many things that will clutter up his house, so values high-quality products or experiences. He does not want clothing. My budget is between $200 and $300. Please provide specific product recommendations, including their prices and where to purchase them.”

How to refine your results?

One of the great things about a generative AI tool like ChatGPT is that it’s conversational. You can ask ChatGPT to help you refine your prompt, and it will suggest specific details to include.

Additionally, working with AI is always an iterative process. When you try out your prompt, you may find that the outputs don’t align with what you were hoping for. For example, I initially tried a different prompt, which ended up generating a list of clothing options for my dad. I adjusted the prompt to be more specific about his interests and clarified that he doesn’t want clothes.

You can go back-and-forth with ChatGPT, refining your prompt multiple times until you achieve exactly the results you’re looking for.

Prompt engineering is your key to unlocking the full potential of generative AI models like ChatGPT, Gemini or Claude — and it’s not just for business! With just a bit of prompt crafting, you can transform a typical generative AI tool into a powerful, time-saving assistant that helps you discover exciting, personalized gifts in a fraction of the time, making it easy to truly win gift giving this holiday season!

Advertisement

Paige Kassalen

Paige Kassalen has an electrical engineering degree from Virginia Tech and a Master of Information Systems Management from Carnegie Mellon. Kassalen began her career as the only American engineer working with Solar Impulse 2, the first solar-powered airplane to circumnavigate the globe. This role landed Kassalen a spot on the 2017 Forbes 30 Under 30 list along with feature articles in Glamour, Teen Vogue, and Fast Company. Since Solar Impulse, Kassalen worked in the manufacturing and finance industries to create implementation strategies for a range of emerging technology trends from autonomous vehicles to machine learning. She was the Chief Operating Officer at CrowdAI, a start-up named by Forbes as one of the most promising AI companies in 2021. CrowdAI was acquired by Saab, Inc. in 2023, and Kassalen now serves as the Chief of Staff for the strategy division.

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button