Senior AI Performance Architect

Technology, Data & Digital · Data, AI & Analytics · Machine Learning · Data Engineering · Software Engineering

In short

Qualcomm is looking for a Senior AI Performance Architect to design and enhance AI training systems, focusing on hardware accelerators, memory, and scale-out architectures. This role involves understanding ML trends, customer requirements, and performing detailed performance/power trade-offs for future AI models.

Responsibilities

Understand ML network design trends and their impact on SW/HW design.
Work with customers to define hardware requirements for AI training systems.
Analyze current accelerator and GPU architectures.
Architect enhancements for efficient AI model training.
Design and architect flexible computational blocks with various datatypes and precisions (e.g., floating point, 1/2/4/8/16/32-bit).
Design and architect memory subsystems optimized for capacity, bandwidth, and compute-in-memory/near-memory.
Design and architect scale-out and scale-up architectures, including switches, NoCs, and communication collectives.
Optimize architectures for power efficiency.
Perform competitive analysis.
Codesign HW with SW/GenAI (LLM) requirements.
Define performance models to validate architecture proposals.
Perform pre-silicon performance prediction for ML training workloads.
Analyze performance/area/power trade-offs for future ML algorithms and SOC components.

Requirements

Master's degree in Computer Science, Engineering, Information Systems, or related field.
3+ years of Hardware Engineering experience defining architecture of GPUs or accelerators for AI model training.
In-depth knowledge of nVidia/AMD GPU capabilities and architectures.
Knowledge of LLM architectures and their HW requirements.

Desired Qualifications

Knowledge of computer architecture, digital circuits, and hardware simulators.
Knowledge of communication protocols used in AI systems.
Knowledge of Network-on-Chip (NoC) designs in System-on-Chip (SoC) designs.
Understanding of various memory technologies used in AI systems.
Experience in modeling hardware and workloads for performance and power estimates.
High-level hardware modeling experience preferred.
Knowledge of AI Training systems such as NVIDIA DGX and NVL72.
Experience training and finetuning LLMs using distributed training frameworks like DeepSpeed, FSDP.
Knowledge of front-end ML frameworks (TensorFlow, PyTorch) for ML model training.
Strong communication skills (written and verbal).
Detail-oriented with strong problem-solving, analytical, and debugging skills.
Demonstrated ability to learn, think, and adapt in a fast-changing environment.
Ability to code in C++ and Python.
Knowledge of various ML model classes (CNN, RNN, etc.).

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.
Flexible Location (Remote)

Skills

Machine LearningHardware EngineeringGPUsLLMComputer ArchitectureDigital CircuitsHardware SimulatorsNetwork-on-Chip (NoC)System-on-Chip (SoC)DeepSpeedFSDPTensorFlowPyTorchC++PythonCNNRNN

#AI#Machine Learning#Architecture#Hardware#Performance#GPU#Accelerator#Training#Deep Learning#Data Center#End Devices#Edge#Cloud