Instead of being a deterrent, having a genuinely hard scientific problem is a powerful recruiting tool. It attracts curious, convention-challenging people who are motivated by solving what others cannot and are willing to work through ambiguity to achieve a breakthrough.

In the 20th century, careers like investment banking thrived on networks ("who you know"). The internet made expertise discoverable, shifting value to "what you know" roles like hedge fund managers and AI engineers. This trend continues, making deep knowledge more valuable than a good rolodex.

The stock market is a 'hyperobject'—a phenomenon too vast and complex to be fully understood through data alone. Top investors navigate it by blending analysis with deep intuition, honed by recognizing patterns from countless low-fidelity signals, similar to ancient Polynesian navigators.

A scientific background can be a major asset in a CEO role, not a liability. The core principles of science—making data-driven, rational, and unemotional decisions—translate directly to the business world. This allows for objective choices that align scientific development with the company's business needs.

Tim Guinness prioritizes recruiting graduates with engineering degrees for investment roles. He believes engineers are uniquely trained to make decisions with incomplete information and can handle complex numerical and statistical analysis, which are critical skills for evaluating companies.

Even in hyper-quantitative fields, relying solely on logical models is a failing strategy. Stanford professor Sandy Pentland notes that traders who observe the behavior of other humans consistently perform better, as this provides context on edge cases and tail risks that equations alone cannot capture.

Analyzing a company's human capital reveals surprising correlations for stock performance. A higher number of PhDs per dollar of market cap is linked to better future returns, while a higher concentration of MBAs acts as a negative indicator.

In the late 90s, Credit Suisse's prop desk gained a significant edge through three key innovations: hiring technologists from IT departments and compensating them like traders, systematically collecting and cleaning data, and applying then-nascent natural language processing to trade on news feeds.

For cutting-edge AI problems, innate curiosity and learning speed ("velocity") are more important than existing domain knowledge. Echoing Karpathy, a candidate with a track record of diving deep into complex topics, regardless of field, will outperform a skilled but less-driven specialist.