Work on Dynamo's open-source GPU-accelerated inference stack, extending disaggregated serving across vLLM, SGLang, and TensorRT-LLM to support agentic inference (long-horizon reasoning, tool calling, stateful multi-turn execution). Responsibilities include inference-state management (KV/prefix caches, memory/storage hierarchies), building distributed frontends, optimizing throughput and latency, and contributing to upstream API compatibility. Requires 10+ years' experience, strong Rust and Python skills, and expertise in LLM inference and GPU optimization.

NVIDIA seeks a Distinguished Engineer to found and lead the AI Safety & Security Engineering team, owning core harness architecture, baseline engineering standards, and the Find/Validate/Patch roadmap. The role requires mentoring founding engineers, bridging research to production, and strong experience (Bachelor's or equivalent and 18+ years) in architecting distributed systems, ML pipelines, or large-scale platforms. Base salary range: 320,000 USD - 488,750 USD.

NVIDIA is hiring a Lead System Software Engineer to work on a new kernel-mode Linux driver for the latest NVIDIA GPUs. The role requires 10+ years of software development experience and direct Linux kernel/device driver experience, including a track record of upstreaming patches and influencing driver architecture. Position is remote/US-based (Santa Clara listed).

NVIDIA seeks a Software Engineer to design, develop, and automate Linux Vulkan and OpenGL graphics software and GPU driver features, maintain and improve driver quality, and collaborate with test automation and cross-functional teams. Requires a Bachelor's (or equivalent), 3+ years developing graphics system-level software for Linux, and 3+ years programming in C and C++. Base salary range provided: 116,000 USD–189,750 USD (annual).

Role to design and implement Rust APIs and libraries for CUDA functionality, build safe abstractions over native CUDA/C++ interfaces, develop and maintain C/C++ components, optimize performance across Rust/C++/GPU boundaries, and own features end-to-end (design, testing, profiling, packaging, release). Requires 8+ years experience and strong skills in Rust, C/C++, FFI, systems and GPU programming.

Join NVIDIA's DGX Cloud team to design and implement APIs, workflow automation, state management, and integrations with Kubernetes and observability tools (Prometheus, OpenTelemetry, Grafana). Requires 8+ years' experience, proficiency in Go/Java/Python, and strong background in cloud infrastructure, container technologies, and high-scale distributed systems.

NVIDIA seeks a Senior Systems Software Engineer to design, develop, and optimize high-performance C++ software and media streaming pipelines (WebRTC, GStreamer, RTP/RTCP, QUIC). The role requires 5+ years of C++ experience, deep OS and multi-threading knowledge on Windows and Linux, and work on GPU/CPU performance, video codecs, and low-latency transport protocols. Location includes Santa Clara and remote options (US).

Position to design and implement compiler backends and IRs (rustc, MIR, MLIR, LLVM, PTX), develop ahead-of-time and JIT workflows, and define compiler-enforced safety models for Rust on NVIDIA GPUs. Requires strong Rust and compiler expertise, CUDA/GPU knowledge, and 5+ years of relevant experience.

Work on GeForce Now streaming technology: develop and optimize C++ streaming software and media pipelines (WebRTC, GStreamer, RTP/RTCP), analyze CPU/memory/networking and scheduling behavior, support Windows/Linux/QNX and embedded platforms, and integrate streaming software with platform and automotive components. Requires a Master’s (or equivalent), 8+ years C++ experience, deep knowledge of real-time protocols (SRTP, DTLS-SRTP, SCTP) and WebRTC architecture.

Join NVIDIA's NVHPC compilers & tools group to analyze HPC applications and improve performance through compiler optimizations, library/runtime evaluation, and porting across CPU and GPU platforms. Requires 5+ years programming experience, strong skills in Fortran/C/C++, parallel programming (OpenACC/OpenMP/MPI/CUDA), performance analysis/tuning, and numerical fundamentals.

Lead technical architect for NVIDIA's client platform services focused on building scalable tooling and automated service delivery across global endpoint fleets. Role requires 12+ years in client platform engineering, mastery of IaC/CaC (e.g., Terraform, Ansible, Chef, Puppet), Git-based CI/CD, proficiency in Python/Go/Rust, and experience building internal tooling, CLI tools, or microservices to automate platform operations. Responsibilities include setting technical standards, mentoring engineers, collaborating with Security/Networking/Product teams, and driving reliability and developer productivity improvements.

Join a senior engineering team to lead cross-stack projects that integrate and validate GCC/LLVM toolchains, build and CI workflows, and performance-analysis techniques. Requires 8+ years of systems software engineering experience, strong C/C++ and Linux skills, hands-on experience with compilers, linkers, profiling, and the ability to produce measurable before-and-after validation and reusable playbooks.

Join NVIDIA's compiler organization to design and optimize compiler software for NVIDIA CPUs used in AI and HPC. Responsibilities include analysing AI workloads, developing compiler optimizations and code generation, collaborating with upstream/open-source projects and engaging in pre-silicon hardware/software co-design. Requires 8+ years' experience, strong C++ skills, compiler development knowledge (GCC/LLVM/MLIR), and expertise with CPU architectures (Arm, RISC-V, x86_64, PowerPC). Locations mentioned: UK, Switzerland, Germany, and remote.

NVIDIA seeks a senior machine learning/graphics engineer to apply AI and deep learning research to productize neural graphics for gaming, AR, and pro-visualization. The role requires expertise in ML/DL, real-time computer graphics, CV, strong programming skills (Python, C++, CUDA), and experience with TensorFlow/PyTorch. Position is US-based (Durham and Santa Clara listed) with remote options; salary ranges provided for levels 5 and 6.

NVIDIA is hiring a Senior Technical Marketing Engineer focused on the CUDA platform. Responsibilities include evaluating CUDA features, authoring tutorials/guides/blogs, delivering workshops/webinars/presentations, and capturing developer feedback to inform the product roadmap. Requires MS/PhD or equivalent, 5+ years developer experience with GPU compute platforms, competency in C++ and Python, and strong presentation/communication skills. Location listed in the US (Santa Clara noted) with remote options; some travel is required. Provided base salary ranges in USD for two levels.

Develop AI networking communication frameworks, production tools, and large-scale benchmarks for data center and supercomputer environments. Requires 5+ years' experience, professional Python development, strong Linux/CLI skills, and experience across distributed/clustered AI systems. Locations include US (multiple states) with remote options; posted salary ranges provided per levels.

Role involves designing and implementing idiomatic Python APIs and bindings on top of C/C++ foundations, developing native components, defining interoperability boundaries between languages (Python, C/C++, Rust), and owning features across design, implementation, testing, profiling, benchmarking, packaging, and maintenance. Requires strong production experience in Python and C++, GPU/heterogeneous programming, and API/runtime design.

NVIDIA seeks a Senior AI/Deep Learning Compiler Engineer to analyze deep learning networks and develop compiler optimization algorithms and code-generation techniques. The role requires strong CUDA and GPU performance debugging skills, experience with compiler technologies (e.g., MLIR, LLVM, XLA, Triton), and proficiency in C/C++ and Python. The position lists a minimum of 3+ years' relevant experience and provides a base salary range of 152,000–241,500 USD per year.

NVIDIA is hiring a Senior Deep Learning Software Engineer (Inference) to design, build, and optimize GPU-accelerated software and open-source inference frameworks for LLMs, multimodal and generative AI. The role involves performance analysis, profiling, multi-GPU scaling, and contributing to libraries such as vLLM, SGLang, FlashInfer using technologies like CUDA, CUTLASS, Triton and NCCL. Requires strong C/C++ skills and 5+ years of relevant experience.

Role involves designing and implementing features for high-performance CPUs and GPUs, working with application teams on compilation problems, and interacting with open source communities. Requires a Bachelor's in CS/ECE or equivalent (MS/PhD preferred) and 6+ years developing production-quality compilers or tools. Strong C/C++ and parallel architecture/optimization knowledge; experience with LLVM or EDG and OpenMP/OpenACC is a plus. Base salary range listed: 184,000 USD - 287,500 USD.