Tech Stack
Tag name is followed by "@" symbol and proficiency level value.
About proficiency levels:
- 1-2 — basic awareness. Minimal hands-on experience, and a rudimentary understanding of the technology's purpose;
- 3-6 — daily use. Comfortable and regular usage, capable of handling common tasks and challenges related to the technology;
- 7-9 — you are an expert, you can teach others, you know all the pitfalls and tricks;
- 10 — exceptional knowledge, comprehensive understanding, and adeptness in all aspects of the technology, including advanced problem-solving. Think twice before claiming or demanding such level.
Algorithms
CUDA @ 6
Data Structures @ 6
GPU @ 6
HPC
- 1-2 — basic awareness. Minimal hands-on experience, and a rudimentary understanding of the technology's purpose;
- 3-6 — daily use. Comfortable and regular usage, capable of handling common tasks and challenges related to the technology;
- 7-9 — you are an expert, you can teach others, you know all the pitfalls and tricks;
- 10 — exceptional knowledge, comprehensive understanding, and adeptness in all aspects of the technology, including advanced problem-solving. Think twice before claiming or demanding such level.
Details
Responsibilities
We are seeking a self-motivated senior engineer for the Aerial Omniverse Digital Twin team. This hire will lead the design and implementation of a real-time, GPU-accelerated propagation engine that predicts how radio signals travel through realistic 3-D environments—producing both per-link channel characterisations and radio maps at the speed required by production RAN stacks.
As a member of NVIDIA's Aerial team, you will architect and implement a GPU ray-tracing engine that operates at two time scales.
- Planning scale: the engine produces volumetric radio maps—coverage, SINR, and best-server maps at multiple resolutions—composable across cells, frequencies, and beam configurations that network operators use to design and optimise deployments.
- Real-time scale: the same engine delivers per-link multipath channel updates at the millisecond cadence that a production RAN stack requires, using adaptive algorithms that exploit temporal coherence to avoid recomputing what hasn't changed.
Laying the foundations for real-time ray tracing on GPU hardware—algorithms, data structures, and multi-GPU scaling strategies that make sub-millisecond propagation updates feasible—is the defining technical challenge of this role.
Requirements
- PhD in computer graphics, high-performance computing, computational electromagnetics, or a closely related field (or equivalent experience).
- 8+ years of relevant experience.
- Hands-on proficiency with CUDA and at least one GPU ray-tracing framework (OptiX, Vulkan RT, Embree), with a track record of writing production-quality GPU code.
- Proficiency in GPU-friendly spatial data structures (BVH, space-filling curves, hash maps) and the ability to reason about memory hierarchy, occupancy, and compute-vs-bandwidth trade-offs at the kernel level.
- Working knowledge of electromagnetic wave propagation phenomena (reflection, transmission, diffraction, scattering) sufficient to implement and validate a propagation engine.
- Impactful publications in GPU ray tracing, real-time rendering, or deterministic propagation modelling.
Ways to stand out from the crowd
- Experience with real-time or near-real-time ray-tracing engines shipping in production systems.
- Prior work on multi-GPU partitioning for ray-tracing or large-scale simulation workloads.
- Familiarity with 3GPP channel models and wireless network planning tools.
- Knowledge of geospatial coordinate systems and tiling schemes.
Benefits
You will also be eligible for equity and benefits.