Role involves designing, implementing, and optimizing high-performance dense linear algebra software on GPUs, providing technical leadership, and collaborating with product and QA teams. Requires an MSc/PhD (or equivalent) in a related field and 5+ years developing and optimizing numerical linear algebra software using C++ and parallel programming (ideally CUDA, MPI, OpenMP, OpenACC, pthreads).

Responsible for designing, implementing, and optimizing high-performance dense linear algebra software on GPUs, leading projects and mentoring engineers. Requires 8+ years experience, a degree in CS/Applied Math or equivalent, strong C++ skills, expertise in kernel generation and JIT, and experience with GPU-related technologies (CUDA) and performance optimization. Base salary range shown: $184,000–$356,500 depending on level.

Work on communication libraries (e.g., NCCL, NVSHMEM, UCX) and co-design HW/SW solutions to accelerate AI and HPC workloads. Responsibilities include designing and implementing communication technologies, building proofs-of-concept, running experiments and simulations of large GPU clusters. Requires M.S./Ph.D. or equivalent, 5+ years' relevant experience, strong C/C++ skills, experience with parallel programming models and communication runtimes, and solid knowledge of Linux, system and network architecture.

NVIDIA seeks a Senior System Software Engineer to design, develop, test, and optimize system software for next-generation ARM-based SoCs and datacenter products. Requires 6+ years of relevant experience in computer architecture or software development, strong skills in OS internals, multicore systems, performance analysis (including linux perf), device drivers, and performance optimization; familiarity with CUDA and HPC is a plus.

Work on design and implementation of GPU-accelerated libraries and SDKs for unstructured sparsity in deep learning and HPC. Responsibilities include developing a C++ system for sparse tensor computations, DSL specifications for sparse storage formats, on-demand code generation, enabling frameworks like Python/PyTorch, and performance evaluation and optimization. Required: BS/MS/PhD (or equivalent), 6+ years developing and optimizing high-performance C++ software, strong Python and CUDA skills, and experience with parallel programming (MPI/OpenMP) and sparse compilers (MLIR or TACO).

Work on GPU-accelerating high-performance database and ETL applications: analyze and optimize data-intensive workloads, collaborate with hardware/software teams to influence future architectures and programming models. Requires a Master's/PhD or equivalent, 5+ years' relevant experience, strong C/C++ and low-level parallel programming (CUDA, OpenMP, MPI, etc.), and expertise in CPU/GPU architecture and memory subsystems.

The role develops scalable cloud-native services and APIs to support exabyte-scale, high-performance GPU training and inference workflows, working closely with product, platform, SRE, and partner teams. Requires strong foundation in distributed systems, algorithms, data structures, and experience operating backend or cloud-native services (Kubernetes and cloud providers such as AWS/GCP/Azure) using languages like Go, Python, Rust, C/C++, or Java.

NVIDIA seeks a Distinguished Engineer to lead architecture, systems, and software strategy for AI-native wireless infrastructure (RAN and core), drive GPU-accelerated prototyping, and engage with industry standards (3GPP, O-RAN) and partners. The role requires 20+ years of industry experience and deep expertise in communications theory, RAN (5G NR) PHY/L1/L2, baseband signal processing, AI/ML, and programming in Python, C/C++, and Matlab. Salary range provided: 320,000–488,750 USD/year.

Design, implement, and optimize scalable high-performance sparse linear algebra software for GPU architectures (sparse BLAS, direct and iterative solvers, preconditioners, AMG). Collaborate with library engineers, QA, product management, and external partners. PhD/MSc preferred; 5+ years of relevant experience in C++ and parallel programming technologies (CUDA, MPI, OpenMP, OpenACC, pthreads). Salary ranges provided in posting.

Work on JAX core design and performance optimizations for NVIDIA products, collaborating with AI researchers and cross-functional teams. Requires BS or equivalent and 6+ years' relevant experience, proficiency in C/C++ and Python, familiarity with ML frameworks (e.g., PyTorch, TensorFlow), and strong knowledge of CPU/GPU architectures and distributed training. Base salary ranges provided: 184,000–356,500 USD/year depending on level.

Senior engineer role on NVIDIA's storage management plane (DGX Cloud). Responsibilities include maintaining/developing Kubernetes operators and a CSI plugin, building a web-based platform to manage distributed storage, and defining/implementing APIs. Requires 8+ years web development experience, proven Kubernetes/operator/CSI experience, Node.js, Python/Bash scripting, and Linux expertise. Location: US (remote option noted). Salary range provided (USD, annual).

NVIDIA seeks a Senior Software Architect for an applied research team building systems-level software to move data between GPUs, nodes, and storage. The role requires 12+ years in systems software and/or networking, expertise in RDMA/InfiniBand/GPUDirect/NVLink, strong C++/Rust skills, and experience with ML system concepts and inference frameworks. Responsibilities include architecting communication libraries, profiling and optimizing data movement, integrating networking into AI stacks, and contributing to open-source projects.

NVIDIA is hiring a Principal Deep Learning Algorithm Engineer to research and implement inference algorithms, profile and optimize agentic LLM workloads to reduce latency and increase throughput, and design scalable datacenter systems. Requires BS/MS/PhD or equivalent and 15+ years in deep learning/systems, strong Python and C++ skills, and knowledge of computer architecture and GPU/parallel computing. Role is listed as US (Santa Clara) with remote options and a base salary range of 272,000–431,250 USD per year.

Lead design and implementation of a unified multi-tier memory/storage layer and KV-cache systems for low-latency, high-throughput LLM inference. Requires 15+ years building large-scale distributed or storage systems, strong C/C++ and Python experience (Rust preferred), deep knowledge of GPU/host memory hierarchies, RDMA/GPUDirect/NVLink/NVMe-oF technologies, and experience integrating with LLM serving engines.

Role to scope, design, and implement high-performance numerical dense linear algebra software on GPUs, provide technical leadership, and work with product teams and customers. Requires a PhD or Master’s (or equivalent experience), 5+ years developing production software with CUDA and C++, strong fundamentals in finite precision arithmetic and numerical linear algebra. Locations listed: Germany, Poland, France (also remote).

Senior role to develop scalable library software and optimize high-performance parallel numerical applications across CPU and GPU (CUDA). Requires 5+ years experience, strong C/C++ and software-development skills, performance tuning, benchmarking, and collaboration across distributed teams. Python and familiarity with CI/CD and profiling tools are desirable.

Role involves understanding, profiling, and optimizing DL training and inference on state-of-the-art hardware; collaborating across research and engineering teams; implementing production-quality software and building automation tools. Requires 5+ years experience, strong DL and computer-architecture background, and programming skills in C++ and Python; GPU programming (CUDA/OpenCL) and LLM framework experience are advantageous.

Lead design and development of an end-to-end 5G/6G baseband/reference stack; implement and optimize baseband functions in C/C++/Python/CUDA for CPU/GPU/DPU systems; diagnose and resolve performance bottlenecks (CPU, GPU, NIC); support system integration, performance testing, demonstrations and customer field trials. Requires ~8+ years industry experience and background in wireless communications/DSP and PHY/MAC/baseband software.