Software Engineer, Parallel Scientific Computing
About the role
Role focused on developing and optimizing core computational kernels for a custom parallel SPU architecture, spanning from mathematical descriptions and high-level reference implementations to parallel kernels and hardware-aware optimizations.
Responsibilities
- Develop high-level reference implementations of scientific applications (e.g. finite-difference time-domain methods, computational fluid dynamics, electromagnetic wave propagation, etc.) from mathematical and algorithmic descriptions.
- Implement and parallelize scientific applications using our proprietary Software Development Kit (SDK) across SPU simulation, emulation, and real-hardware environments.
- Develop and optimize performance critical kernels for the SPU architecture.
- Own iterative performance optimization loops, benchmark, profile, identify bottlenecks, implement improvements and repeat, until we reach our performance targets.
- Understand applications across abstraction layers, from their mathematical formulation to their mapping onto parallel hardware.
- Identify opportunities to improve our compilation flow, runtime, hardware utilization, and overall system efficiency.
- Collaborate with architecture and performance teams to analyze bottlenecks and co-design innovative software and hardware solutions.
- Independently identify problems and opportunities, propose next steps, and drive work forward without requiring detailed task by task direction.
- Write clear, concise technical documentation for Vorticity’s engineers and customers.
- Stay current with parallel programming models, High Performance Computing (HPC) architectures, numerical methods, and techniques for mapping scientific workloads onto parallel hardware.
Qualifications
- Bachelor's degree in Computer Science, Electrical Engineering, Applied Mathematics, Physics, or related field.
- Master's or PhD in Computer Science, Electrical Engineering, Applied Mathematics, Physics, or related field.
- 5+ years of experience in modern C++ (CUDA C++ experience strongly preferred) for parallel programming and high-performance computing.
- Ability to understand numerical algorithms and translate mathematical descriptions into working implementations.
- Ability and willingness to debug across abstraction layers, from an application or numerical algorithm down through software and into the underlying architecture.
- Strong understanding of computer architecture and the interaction between software and hardware.
- Proficiency with C++, Python, and Linux development environments.
- Excellent written and verbal communication skills.
- Strong ability to work independently and in a team, while taking ownership of ambiguous problems, and determining what needs to be done next.
- Willingness to put in the hard work needed to bring our SPU to life.
- Zero ego.
Skills
About Vorticity
The Fastest Scientific Computing Platform on the Planet
Many pressing problems facing humanity can be solved with faster scientific compute. Currently, it takes billions of dollars to design and develop a new fusion reactor, a hyper-sonic airplane or a new cancer treatment. This is because scientists and engineers have to build very expensive things before they even know if their ideas work. But what if we can simulate the workings of these ideas (and more) on computers first and have high confidence that it works before we build really expensive things? The work of humanity's greatest scientists over the years has given us the physics and math to do this. The problem is that we simply do not have the computing power to solve these big problems. To address this, the Vorticity Inc is re-imagining scientific computing all the way down to processor architecture and system design. Our proven technology is already helping the energy, life sciences and aerospace industries.