CUDA / C++ Developer

We are looking for a CUDA backend developer to join the ACE3 project team. You will design and optimize GPU-intensive backend modules, collaborating with C++ systems engineers and researchers to deliver high-performance simulation and AI workloads. ResponsibilitiesDesign and implement GPU kernels in CUDA C, focusing on: Kernel fusion On-device operations GPU memory optimization Build and use profiling tools (e.g., Nsight) to measure and improve GPU utilization, inference latency, and training throughput. Optimize custom models for deployment with TensorRT or similar inference engines. Work with backend APIs and orchestration layers to expose GPU capabilities. Collaborate with research teams to translate models into performant CUDA implementations. We are also looking for a C++ backend systems developer to join the ACE3 project team. You will focus on high-performance backend engineering, resource allocation, and distributed execution, working alongside CUDA developers and researchers. ResponsibilitiesDesign and implement backend modules in modern C++ (11/14/17), focusing on: Asynchronous I/O utilities Multi-objective resource allocation Distributed execution support Build and use profiling tools to optimize CPU and memory utilization. Develop APIs and backend services that expose system capabilities to orchestration/frontends. Collaborate with CUDA engineers to integrate GPU capabilities into production pipelines. Support scaling workloads across nodes in cloud or enterprise clusters.

Strong C++ (11/14/17) and CUDA C programming experience. Proven experience using GPUs for compute-intensive applications in production. Hands-on with CUDA profiling, debugging, and kernel optimization. Experience in multi-threaded/multi-process architectures and Linux systems development. Proficiency with low-level I/O, memory management, and performance tuning.