Applied Math Libraries Engineer; SPU

Vorticity is building the world’s first Scientific Processing Unit (SPU), a new class of silicon purpose‑built to accelerate scientific computing beyond the limits of GPUs. We are designing tightly coupled software–hardware systems around applied mathematics workloads to deliver order‑of‑magnitude performance gains. Unlocking its full potential requires early, deep engagement from applied mathematics‑driven software engineers who can translate real‑world scientific workloads into executable models, kernels, libraries, and applications that inform both architecture and tooling decisions.

As a Applied Math Libraries Engineer, you will work at the intersection of applied mathematics, scientific computing, and low‑level software. From day one, you will help build the numerical software foundation of the SPU. This role is focused on building reusable mathematical primitives rather than full end‑to‑end scientific applications. The ideal candidate is excited to implement mathematics on a new architecture.

You will work closely with hardware architects, compiler engineers, and runtime engineers to shape how numerical algorithms are expressed, executed, and optimized on the SPU. This position is ideal for someone who enjoys moving fluidly between applied math, numerical algorithms, and low‑level software, and who wants to help build a new scientific computing platform from the ground up.

• Prototype and implement core numerical linear algebra kernels and libraries for the SPU.
• Translate mathematical algorithms into executable, performance‑relevant software.
• Write C, C++, and Python reference implementations to guide hardware, compiler, and runtime decisions.
• Design benchmarks, correctness tests, numerical accuracy tests, and performance models for numerical libraries and scientific workloads.
• Collaborate with hardware architects, compiler engineers, and runtime teams to evaluate algorithm–architecture tradeoffs and ensure numerical primitives map cleanly to the SPU programming model.
• Iterate based on hardware evolution, compiler behavior, benchmark results, and performance insights.

• Strong foundation in applied mathematics, numerical linear algebra, and scientific computing, with the ability to turn mathematical ideas into correct and efficient software.
• Strong proficiency in C, C++, and Python.
• Comfort working close to hardware and writing performance‑critical, low‑level code.
• Experience implementing numerical algorithms yourself, rather than only using existing libraries.
• Ability to reason about memory layouts, cache behavior, bandwidth, arithmetic intensity, and parallel execution.
• Experience with parallel or accelerator programming models such as CUDA, OpenMP, MPI, SYCL, HIP, or similar.
• Solid understanding of concurrency fundamentals, including race conditions, atomics, synchronization, and thread/process behavior.
• Experience working with low‑level GPU assembly, such as NVIDIA SASS, or equivalent native accelerator instruction sets.

• Familiarity with numerical computing libraries such as BLAS, LAPACK, FFTW, Eigen, Suite Sparse, PETSc, cuBLAS, cuSOLVER, cuSPARSE, cuFFT, or similar.
• Experience building numerical libraries, solvers, scientific computing frameworks, or HPC infrastructure.
• Familiarity with performance analysis tools or modeling techniques, including profilers, roofline models, hardware counters, or analytical performance models.
• Exposure to compilers, runtimes, code generation frameworks, or domain‑specific languages for numerical computing.
• Experience applying numerical methods in scientific domains such as physics, geophysics, CFD, climate, materials, fusion, or finance.

• Excellent written and verbal communication skills
• Strong ability to work independently and collaboratively in a team.
• Comfort operating in an early‑stage environment where the hardware, compiler, and software stack are evolving together.
• Willingness to put in the hard work needed to bring the SPU to life.
• Above all: low ego.