Principal SoC Architect, AI Data Center

Technology, Data & Digital · Software & Web Development · Software Engineering · Systems Engineering

In short

Qualcomm's Data Center AI Architecture team is seeking Principal SoC Architects to define next-generation cloud AI SoCs and systems for generative AI inference. This role involves technical leadership, translating requirements into solutions, and quantifying tradeoffs for compute, memory, and system scalability.

Responsibilities

Provide technical leadership across SoC and system architecture studies, product definition, and implementation guidance.
Translate product and workload requirements into scalable SoC and system solutions.
Develop SoC and system models and specifications, and drive feature-set requirements into technology, IP, platform, and software teams.
Quantify system-level performance, power, area, cost, capacity, memory bandwidth, network topology/bandwidth/latency, and scalability tradeoffs.
Follow designs through implementation and production, correlate observed system performance with models and targets, identify root causes of gaps, and drive performance to closure.
Drive competitive analysis across SoC, platform, system, compute, memory, data center networking, software, and deployment considerations.
Partner across architecture, design, systems, product, software, platform, and customer teams to influence roadmaps and deliver market-leading data center systems.
Identify areas for innovations and drive development of patents.
Provide team leadership and mentor technical contributors.

Requirements

Bachelor's degree in Electrical Engineering, Computer Science, or related field and 8+ years of Systems Engineering or related work experience.
Master's degree in Electrical Engineering, Computer Science, or related field and 7+ years of Systems Engineering or related work experience.
PhD in Electrical Engineering, Computer Science, or related field and 6+ years of Systems Engineering or related work experience.
2+ years of experience in one or more system architecture technology areas and products (e.g., Power System, Shared Resource Management, Limits/Thermal Management, Hardware Islands).

Desired Qualifications

MS or PhD in EE, ECE, CE, CS, or a related field.
10+ years of relevant experience in SoC, platform, or system architecture; AI accelerators; memory architecture; or memory technologies.
Demonstrated ownership of complex architectural deliverables, with the ability to lead across teams, build alignment, and influence technical direction.
Deep computer and system architecture expertise spanning processors, accelerators, memory systems, data center network fabrics, interconnects, hardware/software interfaces, platform topology, and SoC and system implementation tradeoffs.
Ability to frame ambiguous problems and lead quantitative SoC, system, and competitive studies using high-level calculators, spreadsheets, analytical models, profilers, and functional or performance simulators.
Experience driving development of architecture methodologies, tools, and analysis frameworks and translating findings into actionable recommendations.
Experience correlating modeled and measured system performance, isolating hardware and software bottlenecks, and driving cross-functional performance closure through production.
Excellent technical and executive communication, documentation, and interpersonal skills.
Self-directed execution, mentorship of others, and rigorous attention to accuracy.
Understanding of generative AI and machine-learning workloads, especially Large Language Model inference and accelerator bottlenecks.
Experience in processor, accelerator, or system architecture, including ISA design, microarchitecture, performance modeling, system topology, scale-up or scale-out interconnects, or hardware/software co-design.
Knowledge of DRAM and memory-controller architecture, bandwidth, latency, and power tradeoffs, protocols such as LPDDR, HBM, DDR, or GDDR, and emerging technologies such as processing-in-memory, processing-near-memory, or 3D integration.
Expertise in one or more data center SoC or system technologies such as RAS, ECC, security, encryption, Ethernet or InfiniBand networking, RDMA, network interface controllers, switches, congestion control, chip-to-chip or die-to-die protocols, chiplets, PHYs, power and thermals, advanced packaging, platform management, or system scalability.
Programming ability for architecture modeling and automation; object-oriented C++ and Python preferred.
Fluent with using AI to accelerate workflows.

Benefits

Competitive annual discretionary bonus program.
Opportunity for annual RSU grants.
Highly competitive benefits package designed to support success at work, at home, and at play.