We are looking for a graphics performance engineer to push Gsplat and other accelerated libraries to speed of light, achieving full system utilization and occupancy and enabling our latest hardware capabilities.
What you'll be doing: Implement, validate, release and maintain highly-optimized reconstruction and rendering algorithms using CUDA and Slang. Optimize data processing pipelines for low latency and maximum throughput. Influence software architecture, validation strategy and technical roadmaps to ensure outstanding performance.
What we need to see: Master's of Science in Computer Science or Electrical engineering (or equivalent experience). 3 years of practical experience. Strong fundamentals in real-time computer graphics. Expertise in GPU-accelerated software with CUDA, Slang, or other shading languages (GLSL, HLSL, Metal) for low-latency, high-throughput applications. Expertise defining and driving performance metrics through profiling and benchmarking. Proficiency with Python and C++. Track record releasing production-grade software. Excellent software engineering fundamentals (source control, CI/CD, testing/validation, packaging, containerization, release). Excellent written, visual, and verbal communication to present performance challenges, tradeoffs, and architectural alternatives. Curiosity and drive to learn new technologies and partner across teams and functions.
Ways to Stand Out from the Crowd: Contributions to 3D game engines, graphics or computer vision SDKs. Algorithmic expertise in neural reconstruction (NERFs, Gaussian Splats). Experience developing high-performance distributed systems. Grounding in mathematical fundamentals such as linear algebra, numerical methods, statistics, and exploratory data analysis. History of multidisciplinary creativity and innovation around performance in multiple problem domains.
XML job scraping automation by YubHub
