Long-form interviews and conversations with tech, AI, and finance leaders (30min+)
This lecture introduces modern language models as industrial-scale systems requiring millions of dollars and trillions of tokens to train, explaining their fundamental operation as auto-regressive next-token predictors that encode language structure through massive statistical modeling.
This lecture introduces First Order Logic as a powerful extension of propositional logic that uses objects, predicates, functions, and quantifiers to compactly represent complex relationships and generalizations without enumerating every possible instance.
This lecture introduces logic as a formal language for knowledge representation and reasoning, contrasting it with probabilistic methods and natural language. It establishes the foundational framework of syntax, semantics, and inference rules, then dives into propositional logic's mechanics including formulas, models, and interpretation functions.
This lecture explains how to learn Bayesian network parameters from fully observed data through simple counting and normalization, while reviewing probabilistic inference methods and d-separation rules for determining conditional independence.
This lecture explains how Bayesian networks compactly represent joint probability distributions through local conditional probabilities, then contrasts inefficient rejection sampling with Gibbs sampling—an MCMC method that iteratively modifies existing samples to satisfy evidence, enabling efficient approximate inference even with rare events.
This lecture transitions from model-free and model-based reinforcement learning to probabilistic reasoning, introducing Bayesian networks as a framework for representing uncertain world states. It establishes probability fundamentals—joint distributions, marginalization, and conditioning—using tensor operations (einops) to provide the mathematical foundation for efficient inference in complex domains.
Andrew Huberman discusses the post-COVID shift toward self-directed healthcare driven by circadian science and supplements, the transformative potential of GLP-1 drugs to potentially eradicate obesity, and the serious risks of unregulated research peptides, while navigating the increasingly polarized political landscape of health information.
Accel's Miles Clements breaks down how to evaluate AI investments using a 'time-to-value versus durability-of-value' framework, defends Cursor's explosive growth against skeptics by highlighting market expansion dynamics, and argues that successful growth investing requires balancing hyper-growth outliers with disciplined vertical SaaS rather than flocking to extremes.
Iran conflict has created unprecedented oil market disruption, cutting 20 million barrels per day through Strait of Hormuz closure—4-5x larger than historic oil crises—triggering massive supply chain risks for semiconductors, fertilizers, and food production that could impact civilization-level infrastructure.
Dr. Alex Marson explains how the convergence of gene editing, AI, and molecular biology enables scientists to reprogram immune cells—particularly through CAR-T therapy—to cure cancers, representing a fundamental shift from treating disease with drugs to engineering living cells with DNA instructions.
Despite US-Iran military escalation and the killing of Iranian leadership, markets remain surprisingly calm with a flat S&P and climbing Treasury yields, suggesting investors price the conflict as short-lived and contained—though the hosts warn this complacency ignores risks of regional destabilization and eroding US global leadership.
Economist Steve Hanke argues that while current oil supply disruptions pose less structural risk than the 1979 crisis due to improved US energy independence and lower oil intensity, the market faces greater danger from an existing stock market bubble and accelerating money supply growth that threatens persistent inflation regardless of oil prices.