Andrej Karpathy might be one of the biggest names in AI, but he has some counterintuitive advice for beginners looking to get into the field.
Andrej Karpathy, the former director of AI at Tesla and a founding member of OpenAI, recently offered invaluable advice to aspiring machine learning enthusiasts. His advice, distilled from his extensive experience teaching Stanford’s renowned CS231n course to his work at the forefront of AI research, is simple yet profound.
“Beginners are often focused on what to do, and I think the focus should be more on how much you do,” he said on the Lex Fridman podcast. “I am a believer in the high-level, 10,000-hour concept, where you have to pick things you can spend time on, care about, and are interested in. You literally have to put in 10,000 hours of work. It doesn’t even matter as much where you put it; you’ll iterate, you’ll improve, and you’ll waste some time,” he added.
“I don’t know if there’s a better way. You need to put in 10,000 hours. But I think it’s actually really nice because I feel there’s some sense of determinism about being an expert at something. If you spend 10,000 hours, you can literally pick an arbitrary thing, and I think if you spend 10,000 hours of deliberate effort and work, you actually will become an expert at it. So, I would focus more on whether you are spending 10,000 hours,” he explained.
Karpathy’s message resonates particularly well in the current AI landscape. With the proliferation of online courses, readily available tools, and the open-source nature of many machine learning projects, it’s easy to get caught up in chasing the newest framework or the hottest research paper. While staying up-to-date is important, Karpathy’s emphasis on dedicated practice highlights a fundamental truth: Mastery requires consistent, focused effort over time.
He acknowledges that the initial choice of focus isn’t as crucial as the dedication applied to it. The key takeaway is to choose something engaging enough to sustain those 10,000 hours. This passion-driven approach allows for organic growth, iteration, and inevitable improvement. He reassures beginners that some wasted time and exploration are natural parts of the process.
Karpathy’s belief in the “determinism” of expertise is empowering. It suggests that proficiency isn’t solely reliant on innate talent but achievable through dedicated practice. This perspective offers encouragement to those intimidated by the complexities of machine learning, reminding them that consistent effort is the most significant factor in their journey to expertise.