Computational Mechanics Scientist (Structural Solver)

Flexcompute is architecting the next generation of multi-physics simulation. We are moving beyond legacy solvers to build a platform defined by speed and physical fidelity. We are seeking a Computational Scientist to lead the development of our new structural mechanics solver.

We prioritize fundamental understanding over tool proficiency. We are looking for an architect who can derive the governing equations from first principles and implement them into a robust numerical framework.

About Flexcompute

We are a “Physics Intelligence” company making hardware innovation as easy as software. Our solvers (Flow360, Tidy3D) are not just faster; they are effectively instant, reducing simulation times from days to minutes using massive GPU parallelism and proprietary algorithms.

• Intellectual Density: We maintain an exceptionally high talent bar. Our team of ~90 includes over 50 PhDs from MIT, Stanford, and top research labs. You will work alongside peers who will challenge your thinking and elevate your work daily.
• Impact: Our technology is currently used to design optical interconnects, 3D chiplets, quantum computing chips, eVTOL aircraft, and next-gen wind turbines. You are building the “math engine” that drives real-world engineering breakthroughs.
• Stability & Growth: We are a Series C stage company with significant funding, revenue-generating products, and a clear path to disrupting the $20B+ simulation market.

Requirements:
• Education: Ph.D. in Applied Mechanics, Physics, Applied Mathematics, or a related field.
• Deep Physics Intuition: You possess an intuitive grasp of continuum mechanics—elasticity, plasticity, contact mechanics, and wave propagation. You can predict how a structure should behave before the simulation runs.
• Advanced Numerical Analysis: You have a theoretical mastery of numerical PDEs. You can mathematically prove the consistency, stability, and convergence of a scheme. You understand the trade-offs between implicit/explicit integration and various spatial discretizations.
• Strong General Programming: Proficiency in at least one high-level language (C++, Python, Julia, Rust, or Fortran) with a focus on algorithmic efficiency. You understand data structures, computational complexity (Big O), and how to write maintainable research code.
• First-Principles Thinking: Demonstrated ability to solve novel problems where no textbook solution exists.

Preferred (But Not Required)

• Interest in learning High-Performance Computing (HPC) and GPU architecture.
• Experience with C++ for scientific computing.
• Exposure to multi-physics coupling (fluid-structure interaction, thermal-stress).

Key Responsibilities

• Solver Formulation: Derive and implement governing equations for structural mechanics, ensuring thermodynamic consistency and physical accuracy across linear and non-linear regimes.
• Numerical Architecture: Design the core numerical framework. You will determine the optimal discretization strategies (FEM, DG, IGA, etc.) to balance stability, accuracy, and computational cost.
• Algorithmic Development: Specific language is secondary to code quality. You will write clean, modular, and efficient algorithms to solve large-scale systems of equations.
• Physics-Driven Validation: Diagnose solver behavior not just as code execution, but as physical modeling. You must distinguish between numerical instability and physical phenomena.

Benefits:
• Collaborative Autonomy: We operate with high trust. You own your module, but you have immediate access to world-class experts in electromagnetics and fluid dynamics for cross-domain problems.
• Continuous Learning: We host regular internal seminars where team members teach each other—from “Advanced Linear Algebra” to “GPU Memory Hierarchies.”
• Compensation: Competitive salary + meaningful equity in a fast-growing startup.
• Well-being: Comprehensive Medical, Dental, and Vision insurance; 401(k); and a gym/fitness allowance.
• Environment: A flat hierarchy where the “best idea wins,” regardless of job title.