Skip to main content

Fortran Developer with background in Math

Fortran 90/95Numerical Linear AlgebraCholesky DecompositionLapack
6 days ago
data_science
L

Luxoft

Kyiv
Work Typeremote
Job Typefull time

About the Position

One of the world's largest providers of products and services to the energy industry has a need to develop and support enterprise information system in Oil & Gas domain. This role focuses on building, integrating, and hardening a Bayesian Optimization (BO)–based engine for optimization in petroleum surface networks.

Responsibilities6

  • Implement and integrate Bayesian Optimization workflows within the existing NEXUS surface network optimization framework
  • Develop and maintain Gaussian Process–based surrogate models used during optimization runs
  • Integrate acquisition functions that support reliable decision-making across challenging objective landscapes
  • Implement initial sampling strategies to ensure stable and repeatable optimization outcomes
  • Contribute to performance tuning, memory efficiency, and parallel execution in large simulation workloads
  • Improve code robustness, testability, and maintainability in a long-lived product codebase

Requirements12

  • Fortran 90/95
  • Experience working with modules, allocatable arrays, and structured Fortran code
  • Ability to modify and extend existing numerical code responsibly
  • Numerical linear algebra fundamentals
  • Practical use of Cholesky decomposition and triangular solves
  • Experience using LAPACK routines
  • Awareness of numerical stability and basic conditioning issues
  • Understanding of Gaussian Process concepts
  • Ability to implement or adapt GP prediction code with guidance
  • Familiarity with acquisition functions such as Expected Improvement (EI) and Upper Confidence Bound (UCB)
  • Understanding of exploration vs. exploitation from a user-outcome perspective
  • Experience implementing or using space-filling designs for initial parameter exploration
Fortran Developer with background in Math
View Original