Ilyes Kerkeni
Ilyes Kerkeni
somewhere between research and trading
Hello, I’m Ilyes. I work in systematic trading and quantitative research. My background spans derivatives pricing, volatility modeling, machine learning, and systematic strategies across both buy-side and sell-side environments.
I’ve previously worked at Goldman Sachs, Crédit Agricole CIB, HSBC, and on independent quantitative research projects ranging from volatility surface construction to algorithmic trading. These days, I spend most of my time thinking about markets, signals, execution, and why models that look great in backtests rarely behave the same way in production.
📍 Currently
- Working on systematic trading strategies.
- Teaching probability and quantitative finance when given the opportunity.
- Reading books faster than I finish them.
❓ FAQ
- What do you do? Mostly think about markets.
- What do you actually do? Mostly debug models.
- What do you really do? Convince myself that this time the backtest is different.
Quant-Assistant Portfolio Manager
Systematic Quant hedge fund (30bn$ AUM)
Working at the intersection of quantitative research and portfolio management.
Quantitative Researcher / Trader
North Coast Capital (via Eruditis)
Developed arbitrage-free vol surface models for equity index options, designed relative value strategies, and built ML-based systematic trading approaches. Also experimented with evolutionary AI frameworks for research automation — before it became a trend.
Quantitative Strategist
Goldman Sachs
FX desk strats. Built and improved pricing models for FX derivatives and exotics — barrier options, forward-starts, vol swaps, variance swaps — working directly with traders on desk requirements.
Quantitative Trader (Freelance)
Eruditis
Built and tested systematic trading strategies using ML and reinforcement learning. Learned that most research papers don’t survive contact with real market data.
Quant-Data Scientist
Crédit Agricole CIB
Worked with the IRD exotic desk and the head of AI at GMD to model and anticipate insurer flows using US rates, vols, and index data. Built models that actually improved performance — which, in quant finance, is rarer than it sounds.
Quantitative Researcher
HSBC Paris
Developed rough volatility and ML-based pricing models (rSig, qGauss) for autocallables under the Europlace Institut de Finance. Spent most of the time making them fast enough to be useful in production.
Quantitative Researcher
Crest, Ensae Paris
Estimated VaR and CoVaR using every method imaginable — parametric, non-parametric, GARCH, Monte Carlo, bootstrapping. Concluded that tail risk is genuinely hard to measure. Everyone already knew this.
M.S. Probabilités et Finance (ex DEA El Karoui)
Écoles Polytechnique — Sorbonne University
- Machine Learning, Neural Networks, Deep Learning, Stochastic control and optimisation, High-frequency finance, Machine learning and optimal trading.
Engineering Degree — Applied Mathematics
ENSTA Paris
- Statistics, Probability, Markov chains, Stochastic calculus, Optimization, Martingales, Time series, Monte-Carlo methods, Machine learning, Mathematical models in finance.
Team Engineer Project — H1N1 & Flu Predictions
Predict vaccination status against H1N1 and seasonal flu using tree-based models (XGBoost, CatBoost, RandomForest) and neural networks.
Regression models to predict rental prices from housing features.
Convolutional neural network written in C to classify images by category.
Algorithmic Trading — Stocks & Crypto
Statistical arbitrage strategies on equities (IBKR API) and crypto (Binance API), implemented in Python.
🎖️ Certifications
💻 Technical Skills
- Languages: Python, C++, C, R, Matlab
- ML/DL: PyTorch, TensorFlow, Keras, Scikit-Learn
- Data: Pandas, NumPy
- Tools: Excel, LaTeX, Dataiku, IBM Watson Studio
⚽ Beyond Markets
- Gym and swimming.
- Tennis 🎾 (lifelong Nadal supporter).
- Reading probability, finance, and biographies.
- Former competitive basketball player 🏀.
- Traveling and collecting airport delays.