A computer science student at the University of Florida is helping to train AI through DataAnnotation Tech. Riley Willis was motivated to take this position after moving out of his parents’ home and noted that it sustains his lifestyle. He mentioned that he’s careful with his expenses; otherwise, data annotation alone wouldn’t be sufficient for him.
Willis spends around 30 hours per week at his computer. By the time Saturday rolls around, he has completed his shifts without having spoken to anyone. As an AI contractor, he works in data annotation for a company called DataAnnotation Tech. His job is completely remote and asynchronous—meaning he doesn’t have a boss, co-workers, or daily meetings. “It’s perfect for an introvert,” he told Business Insider.
His daily tasks generally involve “fact-checking” the outputs of various AI models, although he only tackles projects that appeal to him. Willis explained that DataAnnotation Tech offers an array of project options, such as enhancing AI’s creativity in writing and coding. “You are literally going through and labeling the information based on a series of criteria that differ between projects,” he said. “I personally work on a lot of factual stuff because I find the pay-to-effort ratio is nice. So, I spend a lot of time fact-checking responses to ensure the models aren’t providing false or misleading information.”
With the increasing demand for more accurate AI models, there has also been a notable rise in the need for data annotators and labelers. According to Willis, workers in this sector either generate or label data for training, performing tasks that involve programming, writing, and research.
Willis was attracted to this career path as he sought to move out of his parents’ house. “I wanted a job that wasn’t customer service or anything like that,” he said. “I was already comfortable with the remote environment since I was doing remote school.” When he started, he earned $20 an hour, a rate that has since increased to about $25. He mentioned that depending on the project, hourly rates can be as high as $30.
Although Willis does not work full-time, he manages to cover his expenses, including tuition and rent in Raleigh, North Carolina. “I can pay rent, and I can pay my way through college,” he stated. “I may not have much leftover for discretionary spending, especially since I’m not clocking in full-time, but it’s enough for me to get by.”
His frugality helps stretch his income, although he candidly admitted he “lives poor.” “If I had an extravagant lifestyle—going out to bars or spending a lot online—I probably couldn’t do this,” he explained. “But as long as I keep it simple, I can manage.”
However, the job has its downsides. Willis noted that the repetitive tasks can sometimes be “mind-numbing.” Unlike typical full-time positions, he only gets paid for the hours he is actively working on tasks. “In a conventional job, you might get paid for breaks or chatting with coworkers. With this role, you’re only compensated for the time spent actually working,” he remarked.
Willis mentioned that he occasionally spends “six or seven hours” on research but is only compensated for “five hours of actual work” because he needs to take breaks now and then. Nevertheless, he believes the benefits of the position outweigh its drawbacks. “I appreciate the work-life balance; it’s remote, and I can work whenever I want,” he said. “If I feel like working at 1 o’clock in the morning, I can.”
For individuals seeking a part-time gig to complement their full-time job, Willis suggested that training AI could be an excellent choice. “I believe this setup is ideal for someone already working full-time,” he said. “If you can add an extra hour or two a day, earning an additional $25 to $50, it’s definitely manageable.”
