Computational Scientist

Job Title

Computational Scientist

Role Summary

Develop scalable numerical solvers, neural operator surrogates, and simulation pipelines for electromagnetic and electrostatic problems on complex 3D IC geometries using multi-node GPU clusters.

Work with a cross-disciplinary team focused on hardware and semiconductor simulation, validation, and ML-accelerated scientific computing.

Experience Level

Mid-level — expects advanced research or engineering experience (PhD-level background required by the employer).

Responsibilities

Deliver high-performance simulation and learned-surrogate components and validate their correctness and scalability.

• Develop and scale MPI+CUDA PDE solvers for electrostatics, charge transport, and electromagnetic field problems on multi-node GPU clusters.
• Tune and extend algebraic multigrid preconditioners, Krylov solvers, and mesh pipelines for performance and correctness at scale.
• Build and train neural operators (e.g., FNO, DeepONet, GNO) as high-fidelity surrogates for PDE solvers.
• Design simulation pipelines for training data: sampling strategies, mesh handling, and checks for physical consistency.
• Validate solvers and surrogates against analytical solutions, published benchmarks, and cross-validation between methods.

Requirements

Must-have technical skills and experience; preferred items listed separately.

• Deep expertise in numerical PDE methods (FEM, FVM, or BEM): weak formulations, quadrature, convergence, and error analysis.
• Strong C++ and CUDA skills: kernel development, memory hierarchy optimization, and multi-GPU programming.
• Multi-node HPC experience: MPI, domain decomposition, collective communication, and strong/weak scaling practice.
• In-depth knowledge of sparse linear algebra: Krylov methods, preconditioning strategies, and multigrid concepts.
• Hands-on experience training and evaluating neural operators and related architectures on PDE datasets.
• Practical understanding of AI-for-Science methodology: dataset design from simulations, OOD generalization, and enforcing physical consistency.

Education Requirements

PhD in computational physics, applied mathematics, computational engineering, or a closely related field.

Nice-to-have / Preferred

• Experience with HYPRE, PETSc, or Trilinos.
• Familiarity with multi-node GPU cluster tooling: NCCL, CUDA-aware MPI, NVLink topologies.
• Published work in neural operators, physics-informed ML, or scientific HPC.
• IC design domain knowledge: device physics, semiconductor materials, or layout data formats.

About the Company

Company: Voltai

Headquarters: Palo Alto, CA, United States

Voltai develops AI-driven world models and agents to design, evaluate, and optimize physical systems—focusing on hardware, electronics, and semiconductors to enable AI-led hardware co-design, performance modeling, and cross-domain optimization.