Principal Data Center AI SoC & System Architect - Qualcomm - Grenoble, France
Originally posted 19 October 2026 by the employer.
About the role
Qualcomm’s Data Center AI Architecture team defines cloud AI system-on-chip (SoC) architectures and complete data center systems for generative AI inference workloads. The role focuses on SoC, memory, and system architecture for AI accelerators.
What you'll do
- 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, driving 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.
- Correlate observed system performance with models and targets through implementation and production to identify root causes and drive closure.
- Drive competitive analysis across compute, memory, data center networking, software, and deployment considerations while developing patents.
What you'll need
- MS or PhD in EE, ECE, CE, CS, or a related field (or equivalent software engineering and technical experience).
- 10 years of relevant experience in SoC, platform, or system architecture, AI accelerators, memory architecture, or memory technologies.
- Deep computer and system architecture expertise spanning processors, accelerators, memory systems, data center network fabrics, interconnects, hardware/software interfaces, and platform topology.
- Experience conducting quantitative studies using analytical models, profilers, and functional or performance simulators.
- Proven track record correlating modeled and measured system performance to isolate hardware and software bottlenecks through production.
- Experience programming in languages such as C, C++, or Python.
Nice to have
- Understanding of generative AI workloads, including Large Language Model inference and accelerator bottlenecks.
- Experience in processor or accelerator architecture, including ISA design, microarchitecture, performance modeling, scale-up or scale-out interconnects, or hardware/software co-design.
- Knowledge of DRAM and memory-controller architecture and protocols such as LPDDR, HBM, DDR, or GDDR, plus emerging technologies like processing-in-memory or 3D integration.
- Expertise in data center technologies including RAS, ECC, security, Ethernet or InfiniBand networking, RDMA, network interface controllers, chiplets, PHYs, or advanced packaging.
- Programming ability for architecture modeling and automation using object-oriented C and Python.
Skills: AI SoC, system architect, AI accelerators, memory architecture, data center
More semiconductor jobs: Qualcomm jobs · AI chip design jobs
This role has been open 11 days — well below the 90-day median for AI/ML Hardware Engineering roles.
AI/ML Hardware Engineering · AI ML Hardware
|
Open roles in category
490
|
Median days open
90 d
|
Category median salary (all employers)
$236k
|
See the full market breakdown ▾Category comparison, skills in demand, and who else is hiring
| Metric | Qualcomm | All employers we track in this specialty (490 roles · 77 employers) |
|---|---|---|
| Open roles in this specialty | 95 | 490 |
| Open roles in the wider Software, Firmware & Systems family | 688 | 6243 · 144 employers |
| Median days open | 121 d | 90 d (+31 d vs this employer) |
| Median salary (USD postings) | — | $236k |
Skills observed across this category: AI SoC, system architect, AI accelerators, memory architecture, data center
Who's hiring in this category
- Qualcomm (this employer) · 95 open roles · median 121 d
- NVIDIA · 88 open roles · median 65 d
- AMD · 32 open roles · median 113 d
- Micron Technology · 20 open roles · median 85 d
- Mobileye · 19 open roles · median 88 d
- Analog Devices · 15 open roles · median 48 d
How we counted: 490 open AI/ML Hardware Engineering (AI ML Hardware) roles from 77 employers tracked in the SemiconductorJobs index, counted 30 Oct 2026. Specialty figures count only roles carrying this exact specialty label, so an employer's related work in neighbouring specialties is not included there — it is counted in the wider Software, Firmware & Systems family row. Figures refresh nightly.