Leveling Up: How to Build Prompt Engineering Skills

Though still an emerging field, prompt engineering allows everyday users to tap into the vast potential of artificial intelligence through carefully crafted instructions. The more detailed prompts are, the more customizable and useful the capabilities of large language models like ChatGPT can become.

At its core, prompt engineering is the art and science of designing prompts to yield useful, relevant, and coherent AI responses. It draws on an understanding of the inner workings of large language models (LLMs) and trial-and-error experimentation. Prompt engineers combine creativity with analytical precision to essentially “program” the AI through the prompts to provide the desired information. A well-engineered prompt guides the model to deliver the desired output, similar to how a computer reads a line of code to get its instructions.

Mastering prompt engineering takes practice and patience. But well-crafted prompts can unlock AI’s potential for everything from content creation to decision support and beyond. As prompt engineering evolves from art to science, best practices and principles are emerging. Education options are expanding for those seeking to level up their skills. There is room for people to position themselves as prompt engineering experts to meet the fast-growing need for this new type of literacy. A prompt can be a question that the user asks, such as “Explain to me how the jet stream works.” Or it can be a command such as “Write a haiku about spring blossoms” or “Write a tweet to introduce a new product feature suited for a target audience in [industry] and [location].” Feedback can be simple phrases about tone or content such as “Less formal,” or “Make it shorter.”

This article explores the basics of prompt engineering and its growth. We’ll uncover resources for mastering this new skill and meet the prompt engineers pioneering new frontiers.

Prompts typically combine conversational language with coding concepts like parameters, examples, and instructions.

Structuring Prompts

Prompt formatting tools like brackets and slashes help structure prompts. Use words in brackets to provide context or indicate a specific type of information you want to be included in the response. For example:

• [Define]:
• [Summarize]:
• [Evaluate]: [Option 1: Solar Energy] [Option 2: Wind Power]
• Leave a blank where certain parts are left open for the model to complete by placing them within brackets as in The most important aspect of [topic] is

Use slashes to separate distinct choices or actions:

• Should I take an umbrella today? Yes / No
• [Compare]: Coffee / Tea
• Explain / Describe [topic]

Prompt Length

A prompt can be just a few words or hundreds of words. But, a longer prompt doesn’t always improve performance.

Emma Strubell, PhD – Carnegie Melon University

Emma Strubell is the Raj Reddy Assistant Professor in the Language Technologies Institute in the School of Computer Science at Carnegie Mellon University and a visiting scientist at the Allen Institute for Artificial Intelligence. She also holds a courtesy faculty appointment in CMU’s Department of Materials Science and Engineering.

Dr. Strubell earned her PhD from UMass Amherst working in the Information Extraction and Synthesis Laboratory with Andrew McCallum. Previously, she earned a BS in computer science with a minor in math from the University of Maine. Her research is at the intersection of natural language processing (NLP) and machine learning, and her broad research objective is bridging the gap between state-of-the-art NLP methods, and the wide variety of users who stand to benefit from that technology, but for whom that technology does not yet work in practice.

Jeffrey Dean, PhD – Google AI

Jeffrey Dean joined Google in mid-1999, and is currently Google’s Chief Scientist, focusing on AI advances for Google DeepMind and Google Research. His areas of focus include machine learning and AI and applications of AI to problems that help billions of people in socially beneficial ways. He has a wide variety of interests, including machine learning, large-scale distributed systems, computer systems performance, compression techniques, information retrieval, application of machine learning to search and other related problems, microprocessor architecture, compiler optimizations, and the development of new products that organize information in new and interesting ways.

In 2011, Dr. Dean co-founded the Google Brain project/team, focused on making progress towards intelligent machines. Since then, his individual work has focused on research, systems, and applications for AI and ML, as well as steering the direction of Google’s broader AI/ML and computer science research community.

Daniela Amodei – President & Co-founder, Anthropic

Daniela Amodei is an experienced professional with a diverse engineering, risk management, recruiting, and communications background. Daniela began her career in 2010 at the IRIS Center, University of Maryland, College Park. In 2011, she became a Fellow at Conservation Through Public Health, where she partnered with the CEO to create a strategic grant-making plan and delivered development training to senior staff.

In 2012, she joined Matt Cartwright for Congress as a Field Director and Deputy Field Director, recruiting over 80 volunteers and personally making 11,000 voter calls in key districts. In 2013, she joined the U.S. House of Representatives as a Scheduling and Communications. That same year, she also joined Stripe as a Risk Manager, Core Operations, User Policy, and Underwriting. During their time at Stripe, she led three teams of a total of 26 people, managing 12 reports directly and managing managers of the remaining 14, with an average employee satisfaction rating of 94 percent. In 2018, she joined OpenAI as the VP of Safety and Policy and Engineering Manager + VP of People. In 2020, she became the President and co-founder of Anthropic, an AI safety and research company.

Some universities offer master’s programs in Natural Language Processing (NLP), computational linguistics, or machine learning, including coursework and research opportunities related to prompt design, language modeling, and interaction with AI systems. Programs like these provide a comprehensive foundation in the principles, methodologies, and applications of NLP, with opportunities to specialize or focus on areas like prompt engineering.

Degree programs specializing in AI or machine learning often cover topics such as deep learning, natural language understanding, and human-AI interaction, which are integral to prompt engineering. These programs equip students with the technical skills and theoretical knowledge required to design, optimize, and evaluate prompts for language models and other AI systems.

Different certificate programs in machine learning or data science are also emerging. These programs often include modules or courses focused on natural language processing, deep learning, and AI ethics, providing foundational knowledge and practical skills applicable to prompt design and optimization.

Brown University’s Applied AI and Data Science Program

Brown University’s applied AI & data science certificate from the School of Professional Studies is a learn-at-your-own pace course. Watch video content and live online master classes taught by Brown faculty. The curriculum includes AI framework, data science tools, and generative AI. Students can access some mock interview sessions to help with a career transition.

• Location: Blended format
• Duration: 12 weeks
• Accreditation: New England Commission of Higher Education (NECHE)

Massachusetts Institute of Technology

MIT’s professional certificate program in machine learning & artificial intelligence guides students through an overview of natural language processing, predictive analytics, deep learning, and understanding algorithms. The course takes place in June, July, and August on MIT’s campus in Cambridge, Massachusetts. The course is designed for professionals who have some experience in a technical area. It’s also ideal for data analysts, managers working with a lot of data, or anyone seeking a deeper understanding and hands-on experience with AI.

There are two required courses. One is “Machine Learning for Big Data and Text Processing: Foundations” (two days) and “Machine Learning for Big Data and Text Processing: Advanced” (three days). There are also 12 electives that are individually priced.

• Location: Cambridge, MA campus and live virtual
• Duration: Each course is 2 to 5 days
• Accreditation: New England Commission of Higher Education (NECHE)

Milwaukee School of Engineering’s Master’s in Machine Learning

Earn a master’s degree in machine learning and AI leadership to be at the forefront of machine learning and artificial intelligence technologies. In only 32 credits, students can become leaders on complex projects and develop solutions for ethical uses of AI.

• Location: Online, synchronous courses
• Duration: 32 credits, full- and part-time options
• Accreditation: Higher Learning Commission

PennState’s Artificial Intelligence Master’s Degree

PennState’s Artificial Intelligence master of professional studies degree through the World Campus is offered 100 percent online. This 33-credit course aims to teach students how to understand design, development, and deployment of AI and machine learning and apply that knowledge across various industries. Designed for working adults, this course allows students to complete weekly assignments at their own pace. Choose from multiple start dates each year and take a semester off if needed.

• Location: Online
• Duration: One to four years
• Accreditation: Middle States Commission on Higher Education

Purdue University’s Machine Learning Certification Course

Purdue’s machine-learning post-graduate program helps students learn the in-demand skills for the use of AI now and into the future. Through hands-on projects, students will learn about ChatGPT, Dalle-E, Midjourney, conversational AI, deep learning, and more.

• Location: Online
• Duration: 44 weeks
• Accreditation: The Higher Learning Commission

Texas McComb’s Post-Graduate Program in AI and Machine Learning

Students will learn the most in-demand skills including Python, ensemble techniques and model tuning, deep learning, computer vision, and natural language processing (NLP). Ensemble techniques or methods use multiple learning algorithms to obtain better results than one algorithm alone.

• Location: Online
• Duration: Six months
• Accreditation: Association to Advance Collegiate Schools of Business (AACSB)

Given the evolving nature of prompt engineering and AI technologies, consider exploring specialized workshops, short courses, or online tutorials focused specifically on prompt design, language modeling, and human-AI interaction. Organizations, conferences, and industry events often host sessions or training programs that delve into advanced topics and emerging trends in prompt engineering and related fields.

While full prompt engineering degrees don’t yet exist, many NLP, AI ethics, and computer science programs offer electives or special topics courses. We’ll likely see more dedicated prompt engineering certificates and degrees emerge in coming years as the field matures. Hands-on experience through online courses, internships, and personal experiments are good options for leveling up prompt engineering skills.

AI Academy’s Master in Prompt Engineering

This course teaches students to design and prototype their own ChatGPT-powered product. Students are guided through a format of on-demand videos and live sessions taught by a Harvard expert Giancula Mauro. The course consists of three hands-on lessons with Mauro, on-demand comprehensive video and text content with exercises, and two peer group work sessions, collaborating with other students under AI Academy team guidance. Earn a certificate of completion after successfully fulfilling the capstone project.

Arizona State University’s AI Foundations: Prompt Engineering Course

This prompt engineering course at ASU was designed by Andrew Maynard, an expert in transformative technologies. In only two hours, students can level up their prompt engineering skills and learn to evaluate prompts and create prompts that maximize the productiveness of ChatGPT. This course does not require traditional engineering skills. Students learn about prompts that use natural language.

Carnegie Melon’s LLMs and Prompt Engineering Course

Carnegie Melon’s Executive and Professional Education through the School of Computer Science is the backdrop for this course on large language models (LLMs) and prompt engineering. The course covers an introduction to LLMs, their strengths and weaknesses, a comparison of existing LLMs, and a deep dive into methods for creating the most successful prompts.

Coursera’s Prompt Engineering Specialization

Instructor Dr. Jules White is the director of Vanderbilt University’s Initiative on the Future of Learning & Generative AI and associate dean of strategic learning programs in the School of Engineering, and a professor of computer science in Vanderbilt’s Department of Computer Science. White created one of the first online classes for prompt engineering and is an award-winning researcher and instructor. After one month at about 10 hours a week, a student can apply generative AI tools in many ways in work, education, and daily life.

Deep Learning’s Short Course ChatGPT Prompt Engineering for Developers

Deep Learning’s AI short course is presented in collaboration with OpenAI, the creators of ChatGPT. In only one hour, beginner to advanced students will learn prompt engineering best practices, discover new ways to use large language models (LLMs), and gain hands-on practice writing and iterating their own prompts on ChatGPT. Students will also learn how to build a custom chatbot. The course presenters say that only a basic understanding of Python is needed for success in this course.

GSD Council’s Prompt Engineering Certificate

There are no prerequisites for the GSDC Prompt Engineering certification. However, some experience or knowledge of engineering principles and practices is recommended. The syllabus contains 14 chapters, and testing is an hour-long exam with 40 multiple choice questions. This is a good leg-up for AI career opportunities and to develop practical skills for real-world applications.

Udacity’s Natural Language Processing Nanodegree

A “nanodegree” program is a project- and skills-based educational program that offers a credential once it is successfully completed. Udacity’s nanodegree programs are built from collaboration with industry leaders like Google, GitHub, and others. Learners in the Udacity nanodegree program for machine learning can learn directly from Google’s Deep Learning experts. There are several courses in this online program that introduce and explore the fundamentals of natural language processing (NLP). Students can earn a certification of completion after about two months of skills development and real-world projects.

Vanessa Salvia

Writer

Vanessa Salvia is an Oregon-based freelance writer and editor with a bachelor’s degree in chemistry. As fun as rigorous studies in math and science were, Vanessa took an independent path and developed a prolific career covering lifestyle and healthcare topics for magazines and newspapers, important industries such as concrete construction and building waterproofing, and even hard science. You can get in touch at Sage Media and Marketing.