Odds on Open

In this episode, hedge fund manager Alix returns to Odds on Open to tackle what he calls the most important problem facing young investors today: the complete loss of analog training skills that created the greatest investors of previous generations. Alix runs a hedge fund that deliberately avoids AI tools for analysts, believing they're "extremely dangerous" because they optimize analysts in ways that sub-optimize fund performance. He breaks down why creativity comes from constraints, not abundance—why getting to consensus faster with AI actually makes you worse at generating alpha. The core issues: Gen X is the last generation trained with analog tools, the "junior-senior problem" where hedge funds realize they don't need junior analysts anymore, the attention span crisis, and why Silicon Valley executives don't let their kids use the products they built. Alix introduces the "10K and a pencil" approach, inspectional reading techniques to extract 80% of a book's value, frameworks from Charlie Munger and Peter Kaufman, Art of War principles applied to predator-prey business dynamics, and why reading trains every critical investment skill—pattern recognition, visualization, reading between the lines, leaps of judgment.

Disclaimer: This is not a financial promotion and must not be seen as advice, but only as an educational piece

Cláudia Quintela has spent 25 years connecting early-stage hedge fund managers with institutional capital. She's worked across FX, macro, and systematic strategies at State Street, UBS, Morgan Stanley, and Blenheim Capital, one of the world's largest commodity managers at its peak.

In 2017, she founded Vibe Advisors, an independent advisory boutique focused exclusively on helping emerging managers—particularly systematic, CTA, and macro funds—navigate the hardest part of launch: raising that first $50 to $200 million when you have limited track record, tight capacity constraints, and institutional investors demanding day-one infrastructure.

She specialises in the messy reality of early-stage fundraising: fee pressure, seed negotiations, managed account structures, positioning for allocators who need to sell your strategy internally, and translating complex quant models into language that gets you through the door. Her client base skews heavily toward liquid macro and model-driven managers.

Today, Cláudia runs a portfolio career: advising fund managers and investors, writing weekly about entrepreneurship and capital raising, hosting webinars on AI tools for investor relations and marketing automation, and speaking on panels about women in finance. She's an advocate for the sisterhood and believes the next generation of emerging managers will look different from the last.

Based in London and originally from Porto, she holds an MSc in Finance from LSE and is a CFA charterholder. She's here to talk about what actually works when you're trying to raise institutional capital at the hardest stage—and how emerging managers can build smarter, leaner operations using the tools that didn't exist when she started.

In November 2025, I hosted a fireside chat at Columbia University with Deepak Gurnani, founder of Versor Investments, a $1.4 billion [1] quantitative hedge fund based in New York with offices in New York and Mumbai. Deepak spent two decades at Investcorp, where he built and led the firm’s hedge fund division.

In 2013, he stepped away to found Versor with a singular goal: to build a research-driven quantitative firm focused on leveraging alternative data. This conversation is a continuation of the story we began on Odds on Open with Nishant Gurnani and DeWayne Louis, two of Versor’s partners. In that episode, we explored the systematic strategies that Versor runs. In this fireside chat, we go upstream to understand how it all began.

We talk about:

- Deepak’s journey from IIT to Citigroup to Investcorp

- How the hedge fund industry looked in the 1990s versus today

- What it really takes to spin out and build a quant firm from scratch

- Why Versor adopted cloud computing and alternative data years before most peers

- How small firms compete with giants like Citadel, Millennium, and Jane Street

- What Deepak looks for when hiring researchers

- Why “value proposition” is the starting point for any new fund

- The mindset required to build something that lasts

Versor LinkedIn Page: https://www.linkedin.com/company/versorinvestments/

Research Repository (“Athenaeum”): https://www.versorinvest.com/athenaeum/

1. Data as of December 31, 2024. AUM as per SEC definition for the purposes of item 5F on the ADV Part 1a. For important disclosures, please visit: https://www.versorinvest.com/terms-and-conditions/

Ethan Kho interviews David Orr, a former professional poker player turned hedge fund manager. They discuss David's journey from poker to founding Militia Capital, his investment philosophy, and the lessons learned from both industries. David shares insights on risk management, the importance of finding asymmetric bets, and the challenges of the hedge fund industry. He also offers advice for aspiring investors and reflects on the future of hedge funds.

Discover how top multi-manager hedge funds like Citadel and Millennium attract the brightest minds, the evolving talent strategies in the industry, and the unique traits that set successful candidates apart. Whether you're an aspiring finance professional or just curious about the inner workings of hedge funds, this episode offers valuable insights into the competitive landscape of financial services.

What’s the difference between prediction markets trading and equities trading? On Odds on Open, the world’s #1 prediction markets trader Domer explains how prediction markets work as a form of information-based trading, where news and signals can arrive at any moment, forcing continuous price discovery and repricing. Unlike stock markets, where returns often depend on long-term growth, valuation multiples, and market beta, prediction market strategy focuses on information timing, news flow, and market reaction to new data. Rather than forecasting final outcomes, traders focus on event-driven trading, short-term price movement, and probability trading, exploiting mispriced probabilities and trading event contracts instead of holding positions to resolution. This approach allows traders to generate expected value (EV) and highlights the difference between active trading vs passive investing.Domer also explains how many participants concentrate on high-volume headline markets, while traders look for prediction market edge in event contracts trading across smaller markets. On platforms like Polymarket and Kalshi, opportunities exist in alternative markets and micro events that are less crowded and prone to pricing errors. By specializing in specific market categories and focusing on liquidity, volume, and time horizon, traders can adjust position sizing and holding periods to match their edge. This approach mirrors quantitative trading and event-driven strategies, where domain knowledge and execution outperform broad speculation.Other subjects discussed...How prediction markets trading focuses on short-term price movement and active trading rather than holding event contracts to resolution.Prediction market strategy is based on exploiting mispriced probabilities to generate expected value (EV).Prediction market edge is most common in micro markets and sub-events with lower liquidity and attention.Event contracts trading rewards traders who identify information-driven repricing before markets adjust.Information-based trading in prediction markets reacts to discrete news rather than continuous market noise.Probability trading requires distinguishing mean reversion from true regime shifts after breaking news.Losses in prediction markets are often caused by crowded trades and poor position sizing, not direction.Position sizing must scale with edge and uncertainty to preserve long-term expected value (EV).Platforms like Polymarket and Kalshi allow large traders to temporarily distort prices.Capital concentration in alternative markets can create opportunity for smaller traders.Long-term success depends on repeatable decision-making rather than individual outcomes.Prediction markets exhibit less random variance than equities because prices move on information.Poker develops risk tolerance and variance management applicable to prediction markets trading.Regulation is likely to limit influenceable event contracts while allowing large markets to grow.

❗ACCESS OPPORTUNITIES EXCLUSIVE TO OOO VIEWERS: https://app.youform.com/forms/e2jpsj4z ❗

How do top hedge funds actually hedge trades? At SIG, traders were often told not to hedge. Most assume elite trading firms hedge every position, but Susquehanna (SIG) built its edge by avoiding hedging when trades had positive expected value (EV). Kris Abdelmessih—who later moved into portfolio management roles in energy-derivatives trading outside SIG—explains the firm’s model: centralized risk, large position size, and no hedging unless residual exposure distorted EV or created excessive P&L variance. He shows how this framework let SIG capture options market making edge, tighten spreads, and outcompete firms that hedged mechanically.This episode is definitely a new treat for viewers, linking options pricing, trade sizing, and execution to concepts used in algorithmic trading, alpha generation, portfolio optimization, multi-strategy hedge funds, pod shops, and even sports betting edge.We also discuss...How early exposure to the Amex options pits revealed structural options market making edge from wide bid-ask spreads and fragmented exchanges.Why observing senior traders shaped his understanding of expected value (EV), risk of ruin, and edge compounding in proprietary trading.How late-1990s retail flow and weak market microstructure created abnormal options edge before competition increased.The shift from easy trading to tougher environments as order flow toxicity rose and spreads tightened.Why measuring growth in trading comes from recognizing past mistakes, skill vs. luck, and the paradox of skill.SIG’s belief that markets are mostly efficient and why uncovering inefficiency requires serious labor, pattern recognition, and deliberate practice.Why traders must assume customer order flow is informed, reinforcing humility and disciplined risk models.How SIG’s structured recruiting and education flywheel accelerated mastery of options pricing and liquidity provision.Lessons from working with elite performers like Jason McCarthy and seeing extreme competitive drive and work capacity.Why outlier PMs excel across domains, linking athletic intensity to portfolio management and hedge fund strategies.How choosing between discretionary and systematic trading requires understanding personal strengths and cognitive style.Why young traders must tune out status games in high-prestige quant trading, pod shops, and multi-strategy hedge funds.How moving from SIG to a Chicago prop trading firm introduced him to backer models, P&L splits, and strict risk budgets.Managing natural gas options risk through inventory risk, vega exposure, and shifting volatility regimes.How designing options-themed games for his kids teaches expected value, open outcry trading, and real-world decision-making.

In this episode, former Tower Research and Flow Traders quant Annanay Kapila breaks down the reality of high frequency trading, HFT strategies, and top automated trading systems. He explains what quant researchers actually do inside elite firms—where world-class math talent competes in a true zero-sum market. Annanay goes deep into market making strategies, latency engineering, alpha generation, and how quant trading teams iterate models nonstop to stay competitive. It’s one of the clearest looks at quantitative finance, execution risk, and the real mechanics behind profitable automated trading systems.Annanay also gives a rare inside look at Tower Research crypto trading during the 2022 crypto winter, breaking down crypto trading strategies, API latency edge, and structural weaknesses across early exchanges. He shares first-hand insights from the FTX collapse explained—how withdrawals froze, how APIs failed under stress, and how liquidity vanished in real time. These experiences shaped how he thinks about exchange design, risk controls, and the future of fintech startups. All of this leads into QFEX, a new platform aiming to bring crypto-style perpetuals to traditional assets with real infrastructure and institutional-grade reliability—what Annanay calls “FTX without the fraud.” We also cover hedge fund strategies, automation, and the next generation of trading platforms.We also talk about:How high frequency trading, low latency trading, and market microstructure really work inside Tower and FlowThe two schools of quant trading: Chicago-style traders vs MIT-style systematic tradingHow firms build market making strategies, order book dynamics models, and short-horizon alpha signalsWhy HFT strategies have low capacity and how firms try to scale into longer-term systematic tradingThe shift from pure spread capture to predictive statistical arbitrage and directional signalsWhy prop firms want to launch hedge fund strategies and fee-based businesses like Citadel and Two SigmaWhat quant researchers actually do: feature engineering, execution algorithms, and avoiding overfitting in machine learning for tradingHow QFEX is built: exchange architecture, matching engines, and API latency for 24/7 automated trading systemsHow Annanay hires: identifying real contributors, evaluating technical depth, and filtering candidates for quantitative finance rolesInside Tower Research crypto: crypto trading strategies, API edge, and managing exchange riskFTX collapse explained: liquidity freezes, failed withdrawals, halted APIs, and real-time risk management failuresThe future of trading: fintech startups, algorithmic trading, perpetual futures, and next-generation investing platforms

Former Citadel MD Michael Watson gives a direct look inside a multi-strategy hedge fund and what it was like working under Ken Griffin. He breaks down Citadel’s pod-shop model, its high-intensity capital allocation frameworks, and how Ken built a concentration of talent across quant trading, quant research, engineering in quant, discretionary equities, Python/C++, data engineering, and alternative data. Michael explains how Ken’s recruiting philosophy and performance expectations turned Citadel into a benchmark for the entire hedge fund industry.He also outlines the reality of day-to-day portfolio management and hedge fund engineering at scale: long hours, translating PM mental models into production systems, and building infrastructure for systematic investing. Michael discusses compensation dynamics, pass-through economics, and the skills required for tech jobs in quant trading, breaking into quant finance, and high-impact quant research roles. He closes with how Hedgineer — an all-in-one technology stack for hedge funds he launched after Citadel — uses generative AI for finance to bring multi-manager-grade technology, research pipelines, and analytics to emerging managers.We also discuss...How Citadel evaluates talent across quant trading, discretionary equities, data engineering, and front-office technology within modern hedge fund infrastructureThe role of engineering in quant and why understanding PM mental models is as important as writing code for consistent edgeDeep technical dives into Python internals, performance bottlenecks, and production-grade data pipelines for researchThe difference between systematic investing and discretionary research, and how engineers support both through advanced risk modelingWhy multi-managers dominate through capital allocation frameworks, diversification, and risk controls in pod-based structuresThe economics behind pass-through compensation, PM slope, and how front-office compensation is structured inside the pod-shop modelHow Citadel organizes research, data, and infrastructure across pods in the pod-shop modelThe benefits and limitations of alternative data and how PMs convert domain expertise into alpha generationThe rise of generative AI for finance and its impact on research, risk, portfolio analytics, and model deploymentHow Hedgineer deploys forward-engineered technology stacks for hedge funds and emerging managersBuilding knowledge graphs, MCP servers, unified data pipelines, and research platforms for portfolio management and risk modelingCareer advice for breaking into quant finance, quant engineering career paths, and building leverage through engineering skillsUnderstanding allocator sophistication vs. multi-manager scale in modern hedge-fund ecosystemsThe tension between consolidation (Citadel/MLP/Baly) and the democratization of quant tooling for smaller managersHow startups and emerging managers use AI, infra modernization, and research automation to compete with large multi-managersWhy tech jobs in quant trading increasingly require both business literacy and deep engineering intuitionLessons from eight years inside Citadel’s operating model, performance expectations, and front-office compensation structuresWhat separates elite PMs and analysts in quant research, idea generation, and alpha generationHow Python for finance careers gives engineers leverage across research, analytics, and production systems

What’s it like being a trader at SIG? At Susquehanna International Group, Todd Simkin has trained some of the world’s best traders using poker strategy, probabilistic games, and decision-making under uncertainty that mirror real-world quantitative trading. In this episode, Todd breaks down how SIG teaches trading interns Bayesian updating, asymmetric information, market microstructure awareness, and communication under pressure. From reading opponents at the poker table to interpreting order flow as a market-making trader, he explains how these game-theoretic models build trading intuition, strengthen probabilistic judgment, and sharpen the edge required for systematic trading and derivatives trading.Todd also dives into the traits that distinguish exceptional traders from simply intelligent ones — humility, truth-seeking, and the ability to update beliefs quickly when new information arrives. He explains how SIG screens for these qualities in interviews, what it’s like working or interning at SIG, and why technical skill alone isn’t enough for options pricing or market-making. Todd breaks down how real meritocracy works inside a flat trading desk, how traders collaborate to refine ideas, and how the best quants learn to think critically, debate openly, and iterate their decision process rather than operate alone.How market-making works at SIG and how traders interpret order flow, liquidity, and real-time signalsWhat Bayesian updating looks like in practice during a live trading sessionHow trading systems, not just individuals, drive performance in modern quantitative tradingThe structure of SIG’s pods and how traders collaborate inside a flat trading deskWhy communication and idea-sharing matter more than hierarchy in quantitative researchInsights into SIG’s interview process, including probabilistic reasoning, game situations, and ambiguity testsWhy SIG prioritizes truth-seeking culture and “attacking ideas, not people” in decision-makingWhat it’s like interning at SIG and how long-term projects reveal real trading aptitudeHow SIG evaluates technical skill vs. judgment vs. adaptability in new hiresWhy systematic trading requires parameter tuning, model monitoring, and rapid belief updatesHow traders combine options pricing, market microstructure, and private information to form an edgeThe role of sports trading, insurance risk, and prediction markets inside SIG’s broader ecosystemHow SIG thinks about risk transfer, volatility events, and pricing uncertaintyWhy Susquehanna moves into new businesses only when it can be best-in-classThe philosophy behind SIG’s expansion into prediction markets (Kalshi, PolyMarket, etc.)The economics of risk indemnification, NIL deals, promotions, and event-driven insuranceHow traders apply game-theoretic optimal reasoning beyond poker — in pricing, hedging, and model design