Senior Research Engineer, SandboxAQ

• PhD or MS in Computational Physics, Mechanical/Chemical Engineering, Computer Science, or equivalent
• 3+ years (including PhD) of hands-on experience in CFD code development, specifically on GPUs
• Scientific Software Expertise: Proven ability to build scalable software on multi-GPU systems using JAX, PyTorch, CUDA, or MPI frameworks
• Numerical Mastery: Full fluency in numerical linear algebra, PDE numerical methods, sparse/dense linear solvers, and immersed boundary methods
• Python & JAX: Excellent Python programming skills, with JAX proficiency as a prerequisite
• Documentation: Ability to manage multiple deliverables and produce clear technical documentation for milestone reviews

Desirable

• You will join a specialized engineering group focused on building AI-driven, physics-based optimization tools for next-generation catalytic reactors. We are a global, tech-focused team that combines expertise in AI, physics, and high-performance computing to solve complex challenges in chemical and materials engineering
• The Engineering Team is looking for a Senior Research Engineer, CFD Algorithm Development to join our team. This role is central to our efforts to: develop and deploy a differentiable Computational Fluid Dynamics (CFD) framework on multi-GPU and multi-node systems to optimize chemical reactor designs
• This person will achieve the following goals:
• Lead the development of differentiable CFD solvers using JAX for non-isothermal and reactive flow simulations on multi-node, multi-GPU hardware
• Optimize reactor scale flow by integrating transport equations for up to 50 chemical species and linking catalyst property predictions with gradient-based optimization
• Deliver scalable scientific software that enables physics-based optimization of complex reactor geometries through immersed boundary methods and parallel linear solvers
• This is a high-impact role at the intersection of HPC and AI. You will have the opportunity to move the needle on real-world energy and chemical challenges, working in an environment that values entrepreneurialism, ownership, and transformative impact
• Develop Differentiable Solvers: Lead the creation of differentiable CFD frameworks in JAX, extending capabilities to complex geometries and non-periodic boundary conditions using immersed boundary methods
• Scalable GPU Engineering: Design, implement, and containerize code (Docker/enroot) to ensure reproducibility and scalability across multi-node GPU HPC environments
• Numerical Optimization: Develop optimized parallel linear solvers (FFT or matrix decompositions) and gradient-based scripts to iteratively modify reactor designs based on catalyst activity
• Validate & Benchmark: Rigorously validate simulation accuracy against benchmark results and fixed-bed reactor configurations for both non-reactive and reactive flow regimes
• Cross-Functional Collaboration: Partner with AI model developers to integrate property predictions and generate comprehensive technical reports summarizing algorithm scalability and differentiability