• Join the Developer Tools team to work on software such as Nsight Systems. • Work with a wide variety of hardware platforms, ranging from small embedded systems, to large multi-GPU and NUMA servers and clusters. • Be the communicator between our Nsight Systems team, various chip and system design teams, and the metrics library team. • Work with internal and external users to understand their software and performance goals, and use that knowledge to influence future chip design. • Communicate across multiple teams to collect and understand the requirements. • Understand how the underlying hardware and software works, and use that knowledge to deliver valuable features to the users. • Participate in all phases of the software life cycle. • Collaborate with team members across multiple time zones in a dynamic, high-energy work environment. • Interact with internal and external users, help them get the maximum value out of our products, and deliver their feedback to the product team.

• Design and build performance analysis tools and workflows for AI training and inference workloads • Understand how AI performance engineers work and translate their needs into scalable, intuitive tooling • Develop integrations between profiling infrastructure and AI frameworks and workflows • Collaborate with performance engineers, hardware architects, and software teams to ensure profiling capabilities align with real-world AI workloads • Identify performance bottlenecks in AI workloads and develop automated approaches to detect and diagnose them

• Define the complete software architecture for rack-scale infrastructure products and services, covering control plane services, infrastructure management, firmware, operating systems, kernel drivers, networking fabrics, accelerator software, and user-mode manageability software. • Use Kubernetes and cloud-native primitives as an infrastructure fabric when appropriate. This includes controllers, operators, reconciliation loops, and open source components. These components can operate safely at rack and fleet scale. • Build open source infrastructure software that can be embraced in different forms, including libraries, services, controllers, operators, and integration APIs for internal deployments and CSP environments. • Bridge hardware and software teams across firmware, BMC, BIOS, boot flows, OS images, drivers, networking, NVLink domains, InfiniBand, GPUs, DPUs, CPUs, and system management interfaces. • Translate forward-looking infrastructure roadmaps into formal software requirements, architecture specifications, and execution plans that align teams across the organization. • Partner directly with hyperscalers, CSPs, enterprise customers, internal component leads, vendors, and business partners to align infrastructure capabilities with real-world deployment and integration needs. • Establish reliability, security, validation, and left-shift strategies that reduce risk before hardware reaches production environments. • Mentor senior engineers and technical leads, raising the engineering bar for large-scale networked systems, foundational software, and rack-scale control plane development. • Make high-quality technical decisions in ambiguous environments, balancing customer needs, schedule, hardware realities, software maintainability, open source adoption, and long-term infrastructure evolution.

• Design and implement innovative features for our GPU performance profiling tools. • Collaborate and interact with other teams within NVIDIA, including the GPU HW team, the compiler team, and well as other developer tools teams. • Apply your own knowledge and skills to build tools that are instantly used by CUDA developers all over the world.