The Feynman Method for AI/ML Career Success: Why "Curiosity-Driven Exploration" Is Your Secret Weapon
How the Nobel Prize-winning physicist's approach to discovery maps perfectly onto thriving in modern data science careers
I was sitting in a Chicago café last week, thinking about what separates the data scientists who skyrocket in their careers from those who stagnate. Then it hit me: the answer was hiding in a book I’d been reading—Richard Feynman’s The Pleasure of Finding Things Out.
And here’s the wild part: Feynman’s approach to physics discovery is almost identical to a cutting-edge technique in machine learning called “curiosity-driven exploration.”
Let me explain why this matters for your career.
The $200K Career Secret Hidden in a Physics Book
Feynman once said something that changed how I think about career growth:
“The first principle is that you must not fool yourself—and you are the easiest person to fool.”
When I got laid off in 2025, I could have fooled myself into thinking I needed weeks to recover. Instead, I was hired within the same week. Not because I’m lucky, but because I’d been practicing what I now call “career curiosity-driven exploration”—constantly venturing into uncomfortable territory, building skills I didn’t need yet, and being honest about my gaps.
Here’s what most AI/ML professionals get wrong: they optimize for their current role instead of their next three roles.
What Reinforcement Learning Teaches Us About Career Growth
In machine learning, there’s a problem called the “exploration-exploitation tradeoff.” An AI agent can either:
Exploit what it already knows works (safe, predictable returns)
Explore new territory (risky, but potentially massive rewards)
The agents that only exploit get stuck in local minima—they find a decent solution and stop there. Sound familiar?
Modern RL researchers solved this with something called curiosity-driven exploration: the agent gets rewarded for discovering things that surprise it. The discrepancy between what it expects and what actually happens becomes a reward signal. The bigger the surprise, the more it explores that territory.
Translation for your career: The discomfort you feel when learning something new isn’t a bug—it’s a feature. That discomfort is your curiosity reward signal.
When I moved from academic research at the University of Chicago to industry at Philips, I felt like an imposter every single day. That feeling? It meant I was exploring exactly the right territory. Within two years, I’d contributed to multiple FDA 510(k) clearances and had 10+ patents filed.
The Feynman Technique (Applied to Your AI/ML Career)
Feynman had a legendary method for learning anything:
Study the concept
Teach it to a child (or write it in simple language)
Identify gaps and go back to the source
Simplify and use analogies
Here’s how I’ve adapted this for career growth:
Step 1: Pick One Thing You Don’t Understand
Right now, think of one concept in your field that you kinda-sorta understand but couldn’t explain clearly. For me recently, it was the intersection of multimodal models and clinical-grade training pipelines.
Step 2: Explain It Publicly
I don’t mean write a PhD thesis. Write a LinkedIn post. Send a Slack message to your team. Ask a “dumb” question in a meeting.
Here’s the career magic: visibility compounds.
Leaders keep mental lists of people who demonstrate curiosity. When a new project lands on their desk, they don’t consult a database—they think of the first name that springs to mind. That name should be yours.
Step 3: Fill the Gaps (Publicly)
When you discover you can’t explain something, go learn it. Then share what you learned. This loop—curiosity → learning → sharing—is the Feynman engine for career growth.
The “Cargo Cult Career” Trap
Feynman famously warned about “Cargo Cult Science”—activities that look like science but lack the intellectual honesty to actually discover truth.
There’s a career equivalent. Cargo Cult Careers look like success on the surface:
✅ Impressive title
✅ Big-name company on LinkedIn
✅ Fancy certifications
But underneath? No genuine curiosity. No uncomfortable growth. No ability to adapt when the landscape shifts.
I’ve seen brilliant PhDs get stuck because they optimized for prestige instead of learning. Meanwhile, a self-taught developer who’s genuinely curious about why transformers work the way they do ends up running the ML team.
The difference isn’t intelligence—it’s intellectual honesty about what you don’t know.
Your Curiosity Audit (Do This Today)
Answer these questions honestly:
What haven’t you learned because it feels “too hard”? (That’s your curiosity reward signal pointing you toward growth)
When was the last time you felt genuinely surprised by something technical? (If it’s been months, you’re over-exploiting and under-exploring)
Can you explain your daily work to a smart 12-year-old? (If not, you might be operating in cargo cult mode)
What would you learn if no one was watching? (That’s where your authentic curiosity lives)
The Practical Playbook
Here’s what I want you to do this week:
Monday: Pick one technical concept you’ve been avoiding. Spend 30 minutes with it.
Wednesday: Share one “dumb question” publicly—in a meeting, on Slack, on LinkedIn. Watch what happens.
Friday: Teach something small you learned to a colleague. Notice what gaps appear when you try to explain it.
This is the Feynman method. This is curiosity-driven exploration. This is how careers compound instead of stagnate.
The Pleasure of Finding Things Out (In Your Career)
Feynman didn’t work on quantum electrodynamics because it was prestigious. He worked on it because it genuinely puzzled and delighted him. That’s why he won the Nobel Prize—not despite the playfulness, but because of it.
Your career works the same way.
The professionals who thrive in AI/ML aren’t the ones who strategically collect credentials. They’re the ones who can’t help but wonder why things work the way they do. They ask questions that make others uncomfortable. They explore regions of the problem space that don’t guarantee returns.
And here’s the beautiful part: when you’re genuinely curious, the work stops feeling like work. You’re not “building your career”—you’re satisfying an itch that won’t leave you alone.
That energy is magnetic. Recruiters notice it. Hiring managers notice it. Your future team notices it.
So what are you curious about?
Teodora Szasz is a Staff-level ML scientist specializing in multimodal clinical AI, with experience spanning FDA-cleared medical devices, academic HPC research, and AI team leadership. She coaches AI/ML professionals at teodora.coach to build careers that compound. Connect on LinkedIn.
P.S. If you felt that discomfort reading this—the “I should probably be doing more of this” feeling—good. That’s your curiosity reward signal. Follow it.
What concept have you been avoiding? Hit reply and tell me. I read every response.