If you want to work for OpenAI, the company behind ChatGPT, a leader of the company shared the skills he was looking for when hiring. Jan Leike, head of supercompatibility at OpenAI, a new team that aims to align the company’s AI systems with human interests, said in a recent podcast episode that his team will be recruiting for a number of research-related positions.
“We’re primarily looking for research engineers, research scientists, and research executives, and I hope we continue to hire a lot of people,” Leike said in an episode of the business-focused Y Combinator-supported nonprofit’s podcast “The 80,000 Hours Podcast.” My guess will be at least 10 before the end of the year. And maybe even more in the years to come.”
Leike says strong candidates should know how to code, understand machine learning, and know how to think critically. He also adds that they must have a passion for making AI safer.
Being a research engineer at OpenAI
Research engineers on OpenAI’s supercompliance team are responsible for designing and implementing experiments for security research and can earn between $245,000 and $450,000 per year, according to OpenAI’s job listing. Their job entails writing code for machine learning training, managing experimental datasets, and fixing bugs.
Leike adds that the job does not require experience in AI security or a PhD in machine learning, adding that this is the “traditional way” to enter the field. Instead, Leike said his team is primarily looking for a candidate who knows how to develop machine learning models.
Being a scientist at OpenAI
Working closely with research engineers, the research scientists are responsible for developing OpenAI’s compliance research goals. A research scientist, who can earn $450,000 a year, is tasked with brainstorming how to tackle problems and guiding how to design research.
Leike says research engineers and scientists require “a lot of critical thinking, asking important questions, and being very curious about the world and technology we’re building.”
Being an administrator at OpenAI
The research manager position oversees research engineers and research scientists. A research manager, who can earn between $420,000 and $500,000 a year, is responsible for determining the direction the team will take.
Leike states that an ideal candidate for a leadership role would have a combination of management experience and machine learning skills. He says ideal candidates can range from someone who has led a research team at a machine learning company, to a manager in a different field who has contributed to large language model projects, to a postdoctoral fellow who has led a small research team.
Leike added that a manager needs to “really care about the problems we’re trying to solve” and be really good at coding and machine learning.
Leike says his team is addressing “one of the most pressing issues” given the risks the technology may pose to humanity, and states that a job candidate who agrees would be a good fit for OpenAI.
Finally, Leike says, “if you want to work in a team of highly motivated, talented people who are trying to solve really tough problems and have lots of resources to do it, this is the place to go.”