Eric Siu explains how he hires top talent for AI teams: he uses Juicebox for sourcing, screens for engineers with multiple promotions, gives real work assignments instead of relying on resumes, and deliberately tells candidates the job is hard and non-9-to-5 to filter for people who can handle it.
I only hire the top 1% of talent because that's all you can afford to do in today's AI world. Here's my process for doing so. Number one, you got to make sure you're using a recruiting sourcing product. So we use Juicebox for this. Let's say you're looking for engineers with at least 2 promotions at 2 different companies. Juicebox helps you parse this out. Plus, you don't have to spend $30,000, $40,000, $50,000, $100,000+ a year just for sourcing. The second piece is making sure that when you're talking to people, that you are talking to them about AI. AI covers all 4 of our core values. Okay, so we want people who are relentless learners. We want people who have a bias to action. We want people who are unreasonably resilient, and we want people who make bold long-term bets. That is all encompassed within AI. Also, what's very important too is, sure, there's something to be said about the resume, maybe an interview really well, but nothing beats the work assignment. If you have them do the work, you can see that's when the rubber meets the road. It's not a good interviewer or someone that has a good resume. It's how they actually go about doing the work. But we want them to mirror the actual work that we need done. I would just say that you want to make sure that you're also anti-selling them too. So towards the end of the meeting with me, I'll say, hey, here are some reasons why you should and join us. Number one, it's not a 9 to 5, so don't expect to clock in, clock out. It's going to be very hard work. Number 2, we want very AI-forward people, people who are very obsessed with AI. We tell them that we also give feedback directly in public. 99.99% of people don't want that. You want to set as many milestones as possible through the interview process because interviewing is hard. You're really just guessing, okay? And you really don't know until maybe 3 to 6 months into it. That being said, if you enjoyed this video, you want more tips on AI and business, just hit follow for more.
