דרושים » תוכנה » Senior LLM Agents Architect

We are seeking a sharp, innovative, and hands-on Architect to help shape the future of LLM inference at scale. Join our dynamic E2E Architecture group, where we build cutting-edge systems powering the next generation of generative AI workloads. In this role, you will work across software and hardware domains to design and optimize inference infrastructure for large language models running on some of the most advanced GPU clusters in the world.

Youll help define how AI models are deployed and scaled in production, driving decisions on everything from memory orchestration and compute scheduling to inter-node communication and system-level optimizations. This is an opportunity to work with top engineers, researchers, and partners across us and leave a mark on the way generative AI reaches real-world applications.

What Youll Be Doing:

Design and evolve scalable architectures for multi-node LLM inference across GPU clusters.

Develop infrastructure to optimize latency, throughput, and cost-efficiency of serving large models in production.

Collaborate with model, systems, compiler, and networking teams to ensure holistic, high-performance solutions.

Prototype novel approaches to KV cache handling, tensor/pipeline parallel execution, and dynamic batching.

Evaluate and integrate new software and hardware technologies relevant to Core Spectrum-X technologies, such as load balancing, telemetry, congestion control, vertical application integration.

Work closely with internal teams and external partners to translate high-level architecture into reliable, high-performance systems.

Author design documents, internal specs, and technical blog posts and contribute to open-source efforts when appropriate.

We have been transforming computer graphics, PC gaming, and accelerated computing for more than 25 years. Its a unique legacy of innovation thats fueled by great technologyand amazing people. Today, were tapping into the unlimited potential of AI to define the next era of computing. An era in which our GPU acts as the brains of computers, robots, and self-driving cars that can understand the world. Doing whats never been done before takes vision, innovation, and the worlds best talent. Youll be immersed in a diverse, supportive environment where everyone is inspired to do their best work. Come join the team and see how you can make a lasting impact on the world.

We are in need of skilled engineers to join our autonomous driving team to invent, execute, and deploy pioneering end-to-end autonomous driving systems. Our strategy has progressed from AI 1.0 constructing a driver from the ground up to AI 2.0 training an intelligent agent to drive. This is achieved by developing LLMs, VLMs, and VLAs to offer exceptional reasoning, planning capabilities, and interaction with the driving system for autonomous driving and general robotics. Lets innovate the future of autonomytogether!

What you will be doing:

Build and train innovative large-scale modelsincluding generative, imitation, and reinforcement learningto improve the planning and reasoning capabilities of our driving systems.

Explore novel data generation and collection strategies to improve diversity and quality of training datasets. Develop, pre-train, and optimize LLM/VLM/VLA models for autonomous driving and robotics applications.

Collaborate cross-functionally to deploy and integrate AI models into vehicle firmware.

Deliver production-quality, safety-critical software that meets performance, safety, and reliability standards.

We are looking for a creative and experienced Senior Firmware Engineer to join our PCIe Firmware teamsomeone passionate about using artificial intelligence to engineer the foundational hardware of the AI revolution.

As an integral part of our team, you'll architect and implement the core of our next-generation devices. This senior role places you at the center of innovation, where you will have a direct impact on our business and technology by solving sophisticated technical challenges. Its a unique opportunity to shape our technology and empower customers to build the supercomputers and AI fabrics of tomorrow.

What You'll Be Doing:
Lead the architectural design, development, and optimization of cutting-edge PCIe firmware, using AI-driven modeling and insights to deliver exceptional performance.

Serve as a trusted technical expert by investigating, debugging, and resolving challenging PCIe firmware issues for our most important customers.

Collaborate closely with our Chip Design, Verification, Software, and Architecture engineers to find root causes and develop robust, long-term solutions.

Champion the integration of AI-assisted diagnostics and generative AI tools across the entire development lifecycle to boost team productivity and innovation.

Translate customer needs and field data into actionable feedback that directly shapes the future of our products.

We are seeking an AI Networking Exploration Architect for our Networking Insights Group to bridge the gap between cutting-edge, hyper-scale AI workloads and the datacenter infrastructure that enables them. You will join a small, focused team of multidisciplinary engineers driving AI workload optimization through deep application understanding and end-to-end systems thinking. Your insights will directly shape our products across the full stackfrom applications and software libraries to hardware architecture and physical design.

What You'll Be Doing:

Model the performance of complex AI workloads to identify bottlenecks and recommend system-level optimizations.

Translate state-of-the-art research into actionable infrastructure, software, and hardware features in partnership with architecture teams.

Rapidly master new AI domains (LLMs, generative models, multimodal systems) and distill key findings for product teams.

Incorporate your deep knowledge of AI applications into our hardware and software roadmaps.

Conduct independent research by formulating hypotheses about workload behavior and validating them through rigorous analysis.

Drive architectural innovation and network optimization by applying your domain expertise to exploratory analysis of real-world Deep Learning (DL) workloads.

Our Networking System Product Engineering team is looking for an excellent Test Design and Validation Engineer. The position requires an understanding of HW and SW to provide stable, efficient, smoothly running production tests, to enable high availability, while ensuring the quality of the products being shipped to customers. The position combines design and hands-on infrastructure development, implementing and maintenance of sophisticated tools that improves development workflow and overall productivity.

What youll be doing:

Design and develop automated tests for networking Switches and Adapters while working closely with the HW, ASIC, and SW Engineering teams to achieve a reliable test with high coverage.

Exposed to various aspects of design, DFT, and test of our next-generation network products.

Take a significant part in the definition and development of tests from the development level to the final test of Network-Systems.

Utilize test suites to find, debug and resolve problems with production process.

Ability to drive projects to full execution in time and working under pressure of schedule and multi project environment.

Supporting and fix bugs of existing code. Supporting production lines for: deployments, patches, maintenance.

Test Verification and Validation (both HW and SW). Troubleshooting and streamlining/optimizing our testing procedures.

Execute tests in different scopes such: regression, performance degradation, functional, and security; report the progress of testing and provide summary reports of the activity.

We are seeking a Technical Marketing Engineer to join our Ethernet Networking team to keep improving our performance leadership in AI. In this pivotal role, you will be the hands-on expert for our Spectrum-X Ethernet platform, showcasing its superiority for emerging AI use cases. You will develop and implement rigorous benchmarks on various GPU clusters, analyzing everything from LLM training to groundbreaking inference workloads. Your primary mission is to translate these performance results into compelling technical content, including white papers, blogs, and presentations, that clearly articulates why our Spectrum-X Ethernet solutions are the definitive choice for modern AI infrastructure.

What you'll be doing:

Design and execute performance benchmarks using industry-standard tools (e.g., MLPerf, UCX, our Collective Communications Library - NCCL and CloudAI) and customer-representative AI workloads on our state-of-the-art GPU clusters.

Translate your benchmark data and technical insights into compelling, high-impact marketing assets and performance-driven sales enablement materials.

Collaborate closely with Product Management, ASIC and Software architecture and Sales teams, provide feedback on product features, and ensure our performance results are technically accurate and impactful.

We are now looking for a Senior Security Validation Architect for Networking AI Platforms! NVIDIA is seeking an outstanding security architect with deep expertise in security validation and advanced testing methodologies to help design, build, and guide secure validation across our next-generation products. The candidate is expected to deliver highly innovative validation and verification solutions that lead to significant industry differentiation and help achieve the long-range goals of the Security organization. Leaders in this role will strengthen NVIDIAs Data Center and networking products against emerging threats while driving adoption of future-proof security validation technologies.

What youll be doing:

Defining and driving implementation and adoption of security validation strategies across NVIDIAs networking platforms security solutionsspanning Silicon, Firmware, and System levels.

Designing frameworks and automation flows to test and verify security functionality during product QA, large-scale manufacturing, and in-field operation.

Collaborating with architecture, design, and product engineering teams to ensure robust verification coverage for hardware roots of trust, secure boot, key provisioning, and cryptographic services.

Driving development of advanced validation and debugging tools capable of detecting complex security anomalies, misconfigurations, and behavioral deviations across systems.

Building and maintaining an end-to-end visibility architecture for security validation metrics, enabling continuous trust across the product lifecycle.

Our New Product Introduction (NPI) team is growing, and We are now looking for a Chip Engineering Operations Program Manager.! The Engineering Operations group has an immediate opening for a top Program Manager. This group is responsible for the silicon schedule from bring-up to production release. We are looking for individuals who thrive in a dynamic environment, and are not afraid to roll-up his their sleeves and get their hands dirty. We desire a self-starter, with great communication skills; as well as a leader, with a deep sense of commitment. You will be responsible for developing wafer forecasts, schedule generation and management, supporting cross-functional teams, and mitigating any programmatic risks. You will also review and improve processes, and handle prototype demand.

Our Networking division is a leading supplier of innovative end-to-end NVLINK, InfiniBand and Ethernet connectivity solutions and services for servers and storage. We offer market-leading solutions that include adapter cards, switches, cables, and software to support networking technologies. Our products optimize data center performance, using our AI solution and deliver industry-leading bandwidth and scalability. In addition, we serve a wide range of sectors, including high-performance computing, enterprise, data centers, cloud computing, big data, and Web 2.0.

What you'll be doing:

You own silicon bring-up schedule from power on to production release.

Negotiate silicon and board demand with teams and drive a bottom-up forecast.

Oversee and manage chip and board allocations across the company.

Lead prototype chip delivery to internal customers.

Track and coordinate engineering deliverables, key milestones and qualification/validation tasks in the new product introduction phase.

Identify and mitigate risks to schedules and programs.

Communicate status to cross-functional teams as well as upper management.

Continuously evaluate internal tools and processes and drive fixes to improve productivity.

Create new and fix existing processes between different teams.

Drive implementation of SW tools for data analytics and process evaluation.

We are searching for Software Architect of Advanced Development to join our growing software architecture Research team. Opportunity for Scientist Graduates. The ideal candidate will be conducting cutting-edge research at the intersection of Networking, Security, and Communications, and working alongside top experts in these fields. With incredible resources in networking, you will be able to impact, contribute and advance these domains for scalable accelerated computing. Topics include but are not limited to remote direct memory access, hardware offloading and hardware acceleration, distributed accelerator networks, AI for networking and security, storage management, cryptography accelerators and architecture. With its unique open culture, We are one of the best industry labs to do Accelerated Computing research.

What youll be doing:

Technically leading the features owns working with customers and R&D on architecture and design of the features.

Clearly define the requirements. research the hardware, firmware, and software existing support and define the solution to match the requirements he defined.

Simulations ranging from specific components to complete data center environments.

Develop SDKs for novel HW capabilities.

Designing and implementing services, runtime systems, and applications over SDK.

Evaluate and optimize application performance.

Partner and collaborate with other forward-thinking team members and external researchers.

Work with intelligent networking machines powered by AI systems that can learn, reason and interact with other network components.

We are looking for an experienced Software Engineer to take part in the design and development of next-generation performance engineering frameworks. In this role, you will focus on Python-based benchmarking and analysis systems, while also contributing to high-performance components in C/C++. Youll work on challenges at the intersection of networking, distributed computing, and AI, building tools that run at scale on clusters, clouds, and data centers.

What you'll be doing:

Design and implement performance benchmarking and analysis frameworks for next-generation AI and HPC workloads.

Work across Python (primary) and C/C++ (for performance-critical modules) to deliver reliable and scalable tools.

Take full technical ownership of our core telemetry engine, including utilizing Jupyter Notebooks and other data analysis frameworks to help performance engineers analyze telemetry results.

Contribute to our DevOps environment, owning the CI/CD pipelines and release processes for your projects.

Drive technical innovation in the performance engineering ecosystem, including taking part in building our next-gen agentic AI assistant.