Kudzu Infotech - Graph Data Scientist - C++/Python

Kudzu Infotech

Technical Experience
• Proven experience (~5 years) in software development with a strong emphasis on simulation, physics engines, or related areas.
• Solid understanding of 3D graphics, real-time rendering, and simulation techniques
• Proficiency in C++, Python, and/or other relevant programming languages
• Experience with technologies, such as CUDA, Pytorch, DGL, PYG, cuGraph and related GPU programming frameworks.
• Familiarity with Agile development methodologies.
• Prior experience in the development of virtual reality (VR) or augmented reality (AR) applications.
• Knowledge of cloud computing technologies and distributed systems

Responsibilities
• Design and implement parallel algorithms and data structures using CUDA C++ programming language to leverage the computational power of GPUs for accelerating scientific simulations, numerical computations, and machine learning algorithms.
• Develop CUDA kernels, device functions, and memory management techniques to optimize performance, throughput, and scalability of GPU-accelerated applications across diverse domains and use cases.
• Collaborate with software architects, researchers, and domain experts to understand project requirements, design specifications, and performance objectives for CUDA-based applications and algorithms.
• Analyze and optimize code performance, memory utilization, and parallel scalability using profiling tools, debuggers, and performance analysis techniques to identify bottlenecks and opportunities for improvement.
• Implement software solutions for parallel computing tasks, including matrix operations, linear algebra routines, signal processing algorithms, image processing filters, and deep learning models using CUDA libraries and frameworks.
• Integrate CUDA-accelerated components into larger software systems, frameworks, or platforms, ensuring compatibility, interoperability, and performance consistency across heterogeneous computing environments.
• Collaborate with hardware engineers, system architects, and platform developers to optimize software-hardware interactions, exploit GPU features, and maximize utilization of GPU resources for compute-intensive tasks.
• Write clean, modular, and maintainable code using software development best practices, version control systems, and coding standards, documenting design decisions, implementation details, and usage guidelines.