Ingénieur Calcul Scientifique – Méthodes Numériques & C++ H/F
Core
Scientific computing engineer to make pricing libraries differentiable via automatic adjoint differentiation (AAD) for a major international financial services firm.
Role type
Scientific computing engineer (C++/HPC)
Builds
Differentiable pricing libraries and numerical algorithms for financial risk modeling
Domain
Financial services + High Performance Computing (HPC)
Deliverable
production ML models
Required skills
C++20 (templates, concepts, move semantics), Automatic Differentiation (AAD), Numerical Analysis (Monte-Carlo, PDEs), Linear Algebra, SIMD vectorization, Python (NumPy, pandas)
Preferred skills
Git, Linux, GCC/Clang, Expression templates, Memory optimization
Technologies
C++20, Python, NumPy, Git, Linux, GCC, Clang
Responsibilities
Analyze and refactor existing numerical chains (EDP, Monte-Carlo, linear systems) for AAD compatibility, Extend internal C++20 AAD library with new operators and derivation graph analysis, Optimize execution using SIMD and reduce component call overhead, Validate industrial code for numerical stability and non-regression, Document numerical noise and stability investigations
Seniority
Junior to Mid-level (with significant internship/research experience)