Chip Huyen is the co-founder of Claypot AI, a real-time machine learning (ML) platform, and the author of best-selling computer science books including: Machine learning system designPublished last May, Machine Learning Overview Interview. She is an adjunct instructor at Stanford University and previously worked at Snorkel AI and Nvidia.
But Huyen is also a member of the committee that runs MLops Learners. MLops Learners is a community of over 12,000 people dedicated to learning and sharing ML production (MLops) best practices, and we also host virtual and in-person events.
So Huen helps out with the group’s Discord community, where he says there’s a lot of discussion about job hunting right now. This is not surprising given recent tech job cuts, including some of the most skilled and sought-after artificial intelligence (AI) and ML talent at both hot startups and big tech.
AI and ML job seekers are on the rise
“I think it’s a little scary for a lot of people,” she said. “We’ve noticed that one of the most popular channels on our Discord right now is career advice.”
Discord posts are anonymous, so participants can share their fears and anxieties privately, she added. “I just hope that we can provide a space for people to express themselves and maybe others will join in as well.”
She noted that even if someone isn’t fired, even if a colleague is, there’s a feeling of “Am I next?”
“It’s a natural instinct to start looking,” she says. “So we’re seeing a shift in the market from an employment standpoint.”
But while the market itself reduces people’s risk, there is currently a lot of uncertainty about what roles to pursue, she added.
“Someone told me recently that they had an offer from home and also an offer from England,” she said. “Two years ago, they would have been very excited to go to a new country and start a job. But now he’s saying if I go to a new country and get fired, I’ll stay in that country. So I see a trend where people are more reluctant to take risks, even if they could get a really good job in a big company overseas.”
What AI and ML job hunters can do now
Huyen emphasized that there are several things job hunters looking for their next AI or ML job can do to land the right position. She said there may be differences depending on the type of company or industry a candidate is applying to, but overall it’s all about making themselves more robust and agile in the face of change.
1. Differentiate yourself.
First of all, think about how you can differentiate yourself from other AI and ML job seekers, Huyen said. “I see a lot of resumes, and a lot of them are exactly the same,” she said. ”[One candidate] In fact, I told us, I’ve spent 4,500 hours on Python. So how do you measure it? But without context, metrics mean nothing. ”
While it’s true that automated resume screening often requires these types of metrics, for startups like Claypot AI, a cookie-cutter resume isn’t enough. She said, “We encourage candidates to be creative with side projects, because we believe there is a lot of value in having interesting ideas and demonstrating creativity in thinking.”
2. Focus on transferable skills.
Non-transferable AI and ML skills are very specific, such as knowing the ins and outs of specific frameworks and tools, Huyen explained. These may not be transferable to other companies (for example, programming languages like COBOL), but are now obsolete. “The scope of our work changes over time, so we like to look for more transferable skills,” Huen said. “So we’re not looking for people who only know one thing, we’re looking for people with a set of skills that can figure it all out: design thinking, knowing how to ask the right questions, knowing how to communicate ideas clearly, being able to figure out what’s going wrong, so when you run into a problem, you’re not just stuck.”
3. Cover data engineering best practices.
In a recent LinkedIn post, Huyen praised the growing role of data engineers. “More and more data scientists are adopting engineering best practices and moving into data engineering (by choice or necessity). Data engineer roles may even be more in demand than data science roles.”
These are great examples of transferable skills, she noted. “I always make the mistake of trying to get better through engineering,” she said. “Machine learning is more specific, but with a good engineering foundation, like systems thinking, you can learn anything.”
4. Consider a generative AI side project.
“I think generative AI is a really exciting area, and I think there are a lot of opportunities to build products on top of them.” [tools]”So if someone’s looking for a project, I highly recommend it. It’s a place where you can get a lot of creativity and not just sit in front of a keyboard and do what you’re told,” Huen said.
She added that this is also an area with a lot of potential. “When the field is saturated, it’s very easy to get discouraged because you feel like anything you can think of, someone else has already done. But in my opinion, this is still a wide-open field.”
