Agustin Lebron - Trading, Crypto, and Adverse Selection

Augustin Lebrun's core thesis for the book The Laws of Trading is that most individuals should not trade, as they likely lack a sustainable edge and fail to factor in hidden risks and costs.

• The "Straussian" takeaway for retail participants is to avoid trading entirely in favor of more accessible paths to financial satisfaction.
• Trading effectively requires an edge superior to the marginal trader; believing one possesses an edge is often a signal of unfactored risks.

Lebrun's career trajectory moved from six years as a chip designer (2002–2008) to quantitative trading, where he worked at Jane Street starting in January 2008, witnessing the 2008 financial crisis from a trading desk.

• He transitioned to consulting for tech growth (management/hiring) after leaving Jane Street.
• Currently, he has launched a new crypto protocol startup focused on cryptographic guarantees against undesirable trading outcomes.

Adverse selection in hiring is systemic because top talent is incentivized to stay at current employers, leaving the open market populated by candidates at the lower margin.

• Employers face a "winner's curse" where the highest bidder often overpays for a candidate whose true value is lower than the market's average bid.
• Hiring outcomes are further skewed because the final selection power lies with the candidate, who can choose from multiple offers, whereas lower-quality candidates have fewer options.
• Lebrun notes that explicit motivation based solely on "wanting to make money" is a red flag in hiring, as it often indicates a lack of understanding of the job's inherent difficulties.

Ideal candidates for trading roles possess a "willingness to grind" and intrinsic enjoyment of mastering difficult, abstract games, rather than extrinsic motivation for wealth.

• Lebrun prioritizes candidates who have reached the top tier (e.g., top 3 in the world) in obscure, non-monetary domains like chess variants, as this signals a capacity for deep focus.
• Trading firms prefer "blank slate" candidates because previous retail trading experience often contains detrimental habits that must be unlearned.
• Training a new trader to become net positive typically requires a period of six to 18 months involving intensive boot camps and iterative decision-making practice.

Lebrun advocates for a "trillion-dollar arbitrage opportunity" in hiring by screening for high general intelligence (g-factor) and grit globally, specifically targeting underserved high school graduates in countries like India and Nigeria.

• He proposes mass screening bootcamps to convert this untapped talent into high-paying roles for Western companies, bypassing the "sheepskin effect" where credentials inflate earnings without skill acquisition.
• In the context of tech hiring, he identifies "brain teasers" and IQ tests (often disguised) as illegal but effective proxies for intelligence, arguing that companies should accept this reality rather than rely on legible but low-signal skill checks like specific coding languages.
• Startups should adopt a barbell strategy: hiring a few elite "A-players" (90th percentile) with high compensation and supporting them with "B-players" (40th percentile) at lower costs, rather than attempting to hire mid-tier talent for high-stakes roles.

Lebrun remains skeptical about the long-term viability of Constant Function Market Makers (CFMMs) like Uniswap, noting that over 50% of liquidity providers lose money due to adverse selection against which fees are static.

• He views the crypto industry as a "shelling point" for experimentation and believes future success will involve integrating crypto's best ideas into traditional finance rather than replacing it.
• He anticipates a wealth transfer within crypto from older Venture Capitalists to younger builders, though he acknowledges the current capital flow involves significant VC-to-VC transfers.
• Lebrun predicts the future of finance will involve the "crypto-ification" of infrastructure (e.g., credentialing via NFTs, cross-border payments), though he doubts the blockchain's current speed makes it suitable for all on-chain transactions.

Lebrun argues that the financial sector's cost of 9% of GDP is a necessary evil for liquidity and price discovery, despite the prevalence of zero-sum competition within the industry.

• He supports stricter regulation on retail leverage and derivatives products (e.g., banning DTFs for retail) but questions the efficacy of complex capital requirement regulations like Basel III, which he views as creating deadweight loss.
• He expects the trading industry to consolidate over the next decade due to economies of scale, though niche inefficiencies will persist for smaller operators.

Software development in finance is optimized for correctness and safety, leading firms like Jane Street to use strongly typed functional languages like OCaml to make impossible states unrepresentable.

• Lebrun asserts that software development is fundamentally a sociological challenge of managing complexity and organizing teams, rather than just a technical one.
• He recommends treating technical debt as non-recourse financial debt for startups (allowing for rapid MVP iteration) but views it as a massive accumulation of "archaeological cruft" in large legacy enterprises like Microsoft.
• The future of trading will see AI gradually assuming more cognitive load, moving from human-centric models to machine-centric ones, though Lebrun remains skeptical that Large Language Models (LLMs) will drive immediate market alpha due to their sample inefficiency and lack of true semantic understanding.

Lebrun's personal philosophy emphasizes "sequential excellence," where individuals should dedicate 6-to-7-year blocks to mastering a single domain before transitioning to a new field.

• He warns against the pressure to diversify early in life, advocating instead for deep expertise that provides a foundation for future cross-domain insights.
• Career outcomes depend on whether an individual is "evolutionary" (pushing boundaries of existing fields) or "revolutionary" (creating entirely new paradigms), with the latter often benefiting from broad, eclectic exposure.