Qualcomm
Senior AI Performance Architect
$127K–$190K
Raleigh, North Carolina +2 · On-site · Senior · Full-time · Posted Oct 23, 2026
- Location
- Raleigh, North Carolina
San Diego, California
Santa Clara, California - Experience
- 3+ years of Hardware Engineering experience defining architecture of GPUs or accelerators for AI model training. Minimum qualifications also state: bachelor's degree and 2+ years of relevant work experience, or master's degree and 1+ year, or a PhD.
- Education
- Master's degree in Computer Science, Engineering, Information Systems, or related field. Minimum qualifications also allow a bachelor's degree or PhD in those fields.
About the job
Architects AI accelerator hardware and systems for efficient, scalable AI model training, analyzing performance, power, and area trade-offs across hardware and software.
What you'll do
- Analyze ML trends, customer requirements, and existing accelerator and GPU architectures to guide hardware design.
- Architect accelerator, memory, and scale-out/scale-up enhancements for efficient AI training, including hardware/software co-design.
- Build performance models and predict workload performance; evaluate performance, area, and power trade-offs.
What they ask for
- Master's degree in Computer Science, Engineering, Information Systems, or a related field.
- 3+ years of Hardware Engineering experience defining architecture of GPUs or accelerators used for AI model training.
- In-depth knowledge of nVidia/AMD GPU capabilities and architectures.
- Knowledge of LLM architectures and their hardware requirements.
Nice to have
- Knowledge of computer architecture, digital circuits, hardware simulators, AI communication protocols, NoC designs, or memory technologies.
- Experience modeling hardware/workloads for performance and power estimates; high-level hardware modeling preferred.
Pay
$126,700.00 - $190,100.00 (base salary)
Tools
nVidia/AMD GPUs, NoC, SoC, C++, Python, DeepSpeed, FSDP, TensorFlow
AI focus
Architects hardware for AI training, including LLM workloads; co-designs hardware with software and GenAI requirements and models performance for ML training workloads.
Model-written summary · the original posting is authoritative