SoC Product Architect, Telecom AI RAN

NVIDIA•Us, CA

About The Position

NVIDIA is transforming the technology landscape, and we're now applying the power of AI to define the next era of computing, with our GPUs at the core of intelligent systems. Our Telecom team is at the forefront of this revolution, building a GPU-accelerated AI-RAN platform to redefine 5G, 5G-Advanced, and 6G radio and baseband design and deployment. As a Lead SoC Product Architect for Telecom AI RAN, you will be instrumental in defining the architecture for radio and distributed unit products and shaping NVIDIA's AI-RAN roadmap. You will collaborate directly with leading operators and Network Equipment Providers (NEPs) to scale software-defined baseband solutions into production.

Requirements

MS in Electrical Engineering, Computer Engineering, Computer Science, or related field, or equivalent experience.
15+ years of experience in wireless RAN/baseband architecture or SoC product definition, with significant focus on 4G/5G L1/L2 and/or O-RAN / vRAN systems.
Proven track record defining SoC or platform architectures for high-performance communications or signal-processing products, including hardware/software partitioning, accelerator strategy, and memory / I/O requirements.
Comprehensive knowledge of 3GPP RAN standards (4G/5G, NR, NR-Advanced) and O-RAN split architectures, including the functions of radio and distributed units, fronthaul, timing, and synchronization.
Understanding of how RAN standards affect compute, latency, and bandwidth needs of radio and distributed units.
Strong knowledge of L1/PHY algorithms (OFDM, MIMO, beamforming, channel estimation, LDPC/FEC).
Understanding of how L1/PHY algorithms run on central processors, graphics processors, and dedicated accelerators.
Practical experience with at least one commercial-grade 4G/5G PHY or MAC implementation.

Nice To Haves

Direct involvement in architecting GPU- or accelerator-based RAN or signal-processing systems.
Experience bringing a major vRAN / O-RAN product from concept to commercial deployment.
Experience defining or implementing AI-native RAN use cases (e.g., AI-assisted scheduling, channel prediction, traffic forecasting, anomaly detection, or energy optimization) and understanding their impact on RU/DU compute and memory.
Hands-on experience with O-RAN and 3GPP standardization activities, ORAN WG contributions, or multi-vendor interoperability programs.
Prior customer-facing role conducting competitive analysis, design-ins, or technical workshops with Tier-1 operators, NEPs, or cloud providers.

Responsibilities

Define the SoC product family and roadmap for Radio Access Networks in telecommunications.
Partition responsibilities across CPU, GPU, and accelerators.
Set power and cost targets for SoC and platform architecture.
Specify scalability from coordinated RU/DU appliances to cloud-hosted O-DU.
Perform detailed workload and use-case analysis for RU and DU (FR1/FR2, MIMO layers, bandwidth, carrier aggregation, CoMP, mMIMO), and translate these into SoC features, memory hierarchies, accelerators, and interconnect specifications.
Drive competitive analysis against traditional baseband ASICs and alternative vRAN / O-RAN solutions, focusing on performance, power efficiency, silicon cost, flexibility, and upgradeability, to refine SoC and system direction.
Collect and synthesize customer and ecosystem requirements from operators, NEPs, and software partners, and integrate them into SoC, board, and software roadmaps with clear, prioritized product requirements.
Collaborate with NVIDIA’s RAN software, AI, and cloud-native engineering teams to ensure SoC architecture, reference platforms, and SDKs effectively support AI-native RAN applications and rapid adoption of new 3GPP/O-RAN releases.
Present deeply technical content to customers, partners, and internal leadership, articulating architectural tradeoffs, performance results, and roadmap proposals for RU and DU products.