Experts often use the analogy of a toddler to describe A.I., suggesting that products like chatbot phenomenon ChatGPT need to be taught everything they know by a real human being.
In their early days, large language models (LLMs) like these are created by developers and programmers who build them up to a usable level. Then comes the point in an A.I.’s lifespan where it needs to learn how to communicate clearly and efficiently.
This is where a new breed of technology employees is being created—and they don’t need to know a thing about coding.
They are the “prompt engineers,” tasked with training LLMs to continuously give users accurate and useful responses.
Despite people in the role raking in six-figure salaries, potential employers often welcome candidates who don’t come from a tech background or have any coding skills. As Tesla’s former head of A.I. Andrei Kaparthy put it: “The hottest new programming language is English.”
The shift in the tech careers landscape comes amid a heated race for the top spot in the A.I. market, which intensified in recent months after OpenAI’s ChatGPT was labeled a game changer.
Google moved to launch Bard, its chatbot competitor, soon after Microsoft revealed Bing was being revamped to incorporate ChatGPT, in which the tech giant is investing $10 billion. Tesla has joined the race with its Tesla Bot, and Chinese search engine giant Baidu is developing its own version, called Ernie Bot.
Prompt engineer postings at the time of writing range from contracted remote work for $200 an hour to full-time positions paying up to $335,000.
One role, advertised by San Francisco–based A.I. research and safety organization Anthropic, asks for basic programming skills.
However, the company emphasized in its job ad that it encourages people to apply “even if [they] do not believe [they] meet every single qualification.”
“We think A.I. systems like the ones we’re building have enormous social and ethical implications,” the company says. “This makes representation even more important, and we strive to include a range of diverse perspectives on our team.”
Peter Pru, CEO of Ecommerce Empire Builders, which designs and grows online sales platforms, has hired and trained two prompt engineers, both of whom are earning six figures a year.
Pru told Fortune that while a “solid” technical background is important, diverse backgrounds are valuable in bringing unique experience to an ever-evolving platform.
Prompt engineers at his Philadelphia-based company, he explained, are tasked with working with teams from design to research and development to create inputs which will result in an effective outcome for clients. They also test new prompts to generate unique responses which can then be fed back to clients to adapt or improve services.
Training more prompt engineers is “absolutely” something the company would consider again in the future, he said, noting that while the training process could be “time-consuming,” he saw it as an investment in success.
“Training typically involves teaching them about A.I. language models like ChatGPT, as well as the intricacies of prompt design and optimization,” he told Fortune.
“With tools like ChatGPT, it’s not just about asking random questions, its more about engineering prompts to get unique responses. Additionally, they need to learn how to analyze prompt performance and collaborate effectively with researchers, designers, editors, and advertising managers.”
The development of prompt engineers as a career path is in its early stages, according to Kieran Scully, manager in the technology division at specialist recruiters Morgan McKinley.
Scully told Fortune that although skills like coding aren’t essential to land the job, candidates with a background in linguistics or critical thinking should at least familiarize themselves with the basics of data science, machine learning and deep thinking—even if just through free online courses.
He was echoed by Krystian Dabrowski, CEO at Polish knowledge assessment platform Testportal, who reiterated that “candidates from non-tech backgrounds can definitely find opportunities in the A.I. industry.”
The aptitudes bosses in the industry look for, he said in an email, are persistence, logical thinking, and curiosity.
Cause-effect thinking is also key, Dabrowski added.
“Candidates who can provide more contextual information when crafting prompts will be more likely to receive comprehensive and accurate answers from the A.I.,” he said. “This ability to think critically and understand the importance of context will be invaluable in the A.I. field.”
A.I.-focused jobs in more niche markets such as health care, finance, gaming, and legal will also continue to grow, Scully believes. While research out of Goldman Sachs shows 300 million jobs will be impacted by generative A.I., Scully believes roles like prompt engineers are “evidence that [A.I] has and will continue to create new jobs, some we haven’t considered yet.”
2025 will be the hiring peak
Employers like Testportal believe A.I. has also opened up the jobs market for those without a traditional tech background, with chatbots becoming increasingly commonplace in a variety of long-established sectors.
Although a role in its infancy, those looking to snap up one of the lucrative gigs—either as graduates or in a career shift—can expect to find a wave of jobs in 2025, experts are predicting.
Stephen Hellier, director of London-based IT recruiter Initi8, told Fortune he predicted peak demand would come in 24 months’ time, as early adopters prove the technology’s worth and jobs become more defined.
But right now, this is a new frontier, he added.
“The job is based on an ability to build prompts and not on prior experience, so the candidate will probably normally do task-based interviews,” Hellier said. “This should ensure we are hiring from diverse backgrounds, hiring on traits and ability rather than prior experience.”