Chengguo Yang

Compensation is probably one of the most discussed topics in quant finance. Every week I get questions like: "What's a fair package for a QD with 3-5 years' experience?" "What’s compensation should I expect if this is my first job change?" "Is compensation in global firms still ahead of local funds?" So I've put together an updated Quant Compensation Guide based on conversations across the market. One thing to note: This is intentionally a benchmark version. Think of it as the market median, not the outliers. As we all know, the most interesting part in Quant is always in the details: Different firms. Different pods. Different geographies. Different PnL profiles. If you're curious about a specific role, company, team, or just want to know what people are really getting paid, feel free to drop me a message. Happy to share what I'm seeing in the market.

2

1 Comment

AI is already inside modern HFT, but not as an LLM making buy/sell decisions. At leading quantitative trading firms, ML and deep learning are increasingly embedded across the trading stack: - Alpha & short-term prediction: extracting signal from order flow, limit-order books and high-frequency market data. - Market making & pricing: estimating fair value, liquidity, fill probability and adverse selection. - Execution: optimizing when, where and how to trade under market-impact and latency constraints. - Risk Management: adapting exposure to volatility, liquidity and changing market regimes. - Quantitative Research: accelerating hypothesis generation, feature discovery, backtesting and model evaluation. - Engineering: AI-assisted research infrastructure, code, data analysis and automation. The interesting part is what happens next. Modern HFT still is Mathematics + Statistics + Market Microstructure + ML + Low-Latency Systems And LLMs? They are far more likely to transform the research and engineering layer than replace the nanosecond trading engine. The real competitive advantage may be the ability to move faster through the entire loop: Data → Research → Model → Backtest → Execution → Feedback → New Research The next generation of HFT researchers may therefore need to understand both the mathematics of markets and the machinery of AI. AI in Quant Finance is no longer a future topic, it is becoming part of the HFT stack. #AI #HFT #QuantFinance #MachineLearning #QuantTrading #AlgorithmicTrading #MarketMicrostructure

5

1 Comment

We are opening enrollment for the Fall 2026 Options School @ Institute of Quantitative Finance (joint with Options Project in Quantitative Analytics Program @ Center of Mathematical Finance) led by Anton Filatov, Head of Derivatives Trading and Head of Options Projects at CMF since 2016. 🌍 100% online | Free | Weekends | September–December 2026 | English --- Introductory Classes: Advanced Backtesting Engine for Options Trading (Project): https://lnkd.in/dB2iunQD Quantitative Options Trading: The Beginning https://lnkd.in/d8erq8aa QA session: https://lnkd.in/dhP82mBf Advanced Options Trading: Gamma: https://lnkd.in/d-crczFh Options Greeks overview: https://lnkd.in/dFChzBdt xVA: https://lnkd.in/dsWXthts --- Introductory Literature: Antonio Castagna, FX Options and Smile Risk: chapters 1.1 and 1.2.: https://lnkd.in/d5qgGn82 Jiles Jewitt: FX Derivatives Trader School: chapter 1: https://lnkd.in/daf6NbYj Richard Flavell: Swaps and Other Derivatives: pp 1-30: https://lnkd.in/dMxMfYFY --- The Options School focuses on the following domains: - Options Trading - Volatility Trading - Options Pricing & Greeks - Quantitative Research - Backtesting - Risk Management & Hedging - Options Strategies - Market Microstructure & Execution - Quantitative Development --- Program Overview The program in this cohort is designed for a quantitative research, trading, risk management and development: - Development of Options Backtester - Pricing & Curve Modeling - Crypto Options Module - Portfolio Risk & Vega Hedging - AI Agent Layer - Study of Options and Derivatives literature - Practitioners-led workshops - Options and Derivatives Interview Preparation --- Project Overview · Backtesting Engine. You write the exchange: book, matcher, wire protocol. · Volatility curve engine. Implied-vol curves built from sparse strikes, wide spreads, stale quotes and missing sides. · Strategy pipeline. A strategy is a specification that runs, not a screenshot of an equity curve: signal, package, sizing in vega and gamma, limits, cost model. · Expected liquidation cost and vega hedging. Spread, own-flow surface impact and residual implied-vol risk, with liquidity expressed as vega capacity per bucket and the argument about where prudent valuation is layered on top of expected cost. · Crypto options. The same stack against a market with no close, inverse coin-margined payoffs, and perpetual funding competing with option-implied carry. · AI Agent layer. Typed tools over the same engines. An agent that cannot be replayed as a list of commands is out of scope. More details in the carousel. --- 💡How to Apply - Submit your CV and Motivation letter - Complete a project challenge (you’ll get it via email if selected from your CV and Motivation letter) - Join a live interview --- 📝 Registration is open until Friday, September 25: https://lnkd.in/d4w46nZY

15

1 Comment

Marco Argenti, the firm's chief information officer, said engineers are now focused on transferring "tribal knowledge" to AI agents. https://bit.ly/3UglmOy

10

1 Comment