kdb+/q Developer and Data Engineer with 10 years of experience in data and pricing, including 5 years in financial markets (OTC operational risk at Société Générale; gas forward pricing on an energy trading desk). Certified KX kdb+ Developer (Level 2) and PyKX (Level 2); specialized in designing real-time tick architectures and market microstructure tools in q. Seeking an internship alongside my Sorbonne Executive Programme within capital markets, crypto, or commodities desks.
🎓 Executive Program in Financial Engineering — Sorbonne Université (Nov 2026 – Jul 2027)
-
Quantitative Finance & Derivatives: Black-Scholes, Heston, Dupire, SVI, interest rate models (HJM), American options (HJB), crypto/DeFi derivatives (AMM), and energy markets.
-
Numerical Methods & ML: Advanced Monte Carlo (GPU, variance reduction, QMC), Machine Learning in Python (PyTorch/TensorFlow), high-dimensional regression (GARCH, Copulas, PCA).
-
Microstructure & HFT: Order book modeling, high-frequency statistics, optimal algorithmic trading.
-
💼 10 years in data & pricing, 5 of them in financial markets: OTC operational risk at Société Générale, gas forward pricing on an energy trading desk
-
🏅 KX certified: q Developer Level 2 · KDB-X Python Level 2 (+ KDB.AI Vector Database and PyKX courses)
-
🔎 Open to an internship / part-time role in capital markets or commodities — Paris, Mon–Thu, from November 2026
Real-time tick platform & dashboard · q · ▶ live demo
kdb+tick written from scratch (feed, TP, RDB, RTE, HDB, gateway) with a live trading dashboard over WebSocket and KX Dashboards.
→ No data loss (TP log replay on restart) · automatic EOD to a partitioned HDB · dashboard queries in O(symbols), not O(ticks)
Day-ahead power price forecasting · Python
German power market (DE-LU) on real data: ridge and gradient boosting vs the naive EPF benchmark, walk-forward back-test, forecasts valued through a battery trading strategy.
→ Gradient boosting halves the naive error (rMAE 0.53) · ridge captures 98% of perfect-foresight PnL: lowest error ≠ highest PnL
Hawkes quote simulator · q
Clustered quote arrivals simulated fully vectorised in q, MLE calibration (O(n) log-likelihood, Nelder–Mead), partitioned HDB and tickerplant replay.
→ ~530k quotes/day · 95% CI coverage of 94% / 96% / 92% for (μ, α, β) over 50 fits
Fixed-income tick stack · q, TorQ
TorQ pipeline fed by government bonds on Ornstein–Uhlenbeck yield curves, with a fixed-income library in q (cash flows, price/yield, duration, DV01) and real-time risk.
→ 12 benchmarks (UST, Bunds, OATs) · qspec tests in CI
Options pricer · Python
European, Asian and American options with Greeks: closed form, Monte Carlo, Longstaff–Schwartz, SVI smile calibration, hedging back-tests; validated against QuantLib.
→ 94.3% of a real AAPL option chain repriced within the bid–ask spread · 233 tests
Hawkes fits checked by the time-rescaling theorem (KS test), autocorrelation and Fano factor · options pricer benchmarked against QuantLib and real market quotes · power forecasts tested walk-forward with leakage tests · test suites run in CI (qspec for q, pytest for Python).
q Developer Level 2 (Mar 2026) · KDB-X Python Level 2 (Mar 2026) · KDB.AI Vector Database (Oct 2024) · Introduction to PyKX (Oct 2024)
kdb+/q · qSQL · as-of joins · IPC · tick architecture · attributes & query optimisation · TorQ · PyKX
Python (pandas, NumPy, pytest, scikit-learn, Prophet) · SQL · R · BigQuery · ETL pipelines · Git · Linux · Bash · MongoDB · Docker
Quant: Black–Scholes · SABR · short-rate models (CIR, OU) · Hawkes processes · maximum likelihood · Monte Carlo · stochastic calculus · vanilla derivatives (swaps, options, forwards)
- M2 Applied Mathematics & Quantitative Finance — Université de Lille, 2016 · thesis: simulation of fractional Brownian motion and Hurst exponent estimation
- Data engineer, fuel distribution group (23 subsidiaries, 150 stations) — sales forecasting engine in Python, daily price & volume time series
- Gas pricing analyst, energy trading — multi-maturity forward pricing models (month / quarter / semester) with Reuters market data, daily P&L and position reporting
📚 Currently reading: Q Tips (Psaris) · Numerical Probability (Pagés) · Algorithmic and High-Frequency Trading (Cartea) · Trades, Quotes and Prices (Bouchaud)