We are seeking a Numerical Methods Research Scientist to join Aqemia’s R&D team, focused on the development, analysis, and optimization of numerical methods for physics-based methods that accelerate our drug discovery platform.
You will design, implement, improve and validate numerical methods for physical chemistry and statistical mechanics applications. This role sits at the intersection of numerical analysis, statistical physics, and high-performance scientific computing. You will work closely with research scientists, applied physicists, computational chemists, AI researchers, and data scientists, contributing directly to the advancement of our core physics engine, a critical component of Aqemia’s drug discovery platform.
What You’ll Do
- Develop, analyze and optimize numerical methods for the computation of binding and solvation free energies, with a focus on numeric aspects of the methods (code optimization and/or algorithmic improvement).
Implement the numerical methods to provide fast and efficient physics-based algorithms such as:
Molecular Density Functional Theory (MDFT) and Classical Density Functional theory (CDFT)
Alchemical Solvation Free Energy (ASFE) methods
Other statistical mechanics-based methods for binding and solvation free energy predictions.
Integration of machine learning and statistical mechanical methods in collaboration with ML specialists
Create and perform method validation and benchmarking studies against experimental or high-accuracy simulation data.
Collaborate with fellow scientists across R&D, Platform, Engineering, Portfolio departments to develop methods and integrate them into production software.
Stay current with scientific literature; contribute to bibliographic reviews and internal knowledge sharing.
Clearly communicate progress through presentations, internal reports, and written documentation.
Note that this is not an ML-focused role.
What We’re Looking For
PhD in Statistical Physics, Theoretical Chemistry, Computational Fluid Dynamics, Computational Mathematics, Numerical Analysis, Mechanical Engineering or any other field that involves large scale computing, numerical methods etc.
Or 6 years industrial experience in method development, numerics and code optimization.Proven experience in numerical method development, implementation and code optimization (for example with experience with numerical methods such as PDE solvers, optimization algorithms, finite element/difference methods), evidenced by open source software packages, scientific publications and/or industrial projects.
Strong foundation in numerical analysis (e.g., PDEs, optimization, discretization methods).
Proficiency in scientific programming in Python and a lower level language such as C++, Fortran and/or with GPU programming.
Ability to rigorously read, implement, and extend algorithms and methods from the literature. With a commitment to scientific rigor and structured problem-solving in method development.
Analytical, collaborative, and solutions-oriented mindset.
Strong coding practices: clean, properly-documented, and tested code (unit testing, documentation, version control, collaboration with Git).
Ability to work as part of a team based in both London and Paris.
Experience with high performance computing and parallelization/vectorization.
Experience developing classical or electronic density functional theory methods.
Experience with applying ML to computational methods.
Background in chemical physics, statistical mechanical or molecular dynamics.
- Familiarity with atomistically modelling proteins or other biochemical systems, or cheminformatics python libraries (RDKit, Pandas, etc).
Experience in a drug discovery environment.
Nice to have
